Martin Habets is a doctoral researcher at the EUI Department of Economics. An environmental economist, his work examines how everyday environmental conditions shape human behaviour.
In his research, that led to the working paper ‘Temperature Variations and Domestic Violence’, Martin investigates the link between temperature fluctuations and cases of domestic violence. His starting hypothesis is that even small variations in temperature can make people more uncomfortable or change their routines, which may influence interactions and contribute to more cases of domestic violence.
“Temperature affects an extraordinary range of outcomes—from productivity to crime," affirms the EUI researcher. He sadly adds that "domestic violence is, in many ways, a perplexing form of harm: It takes place within relationships meant to offer care and safety, yet it remains one of the most widespread forms of violence worldwide." "The combination of its scale, its emotional complexity, and the lack of evidence on how environmental factors influence it, led me to explore whether temperature plays a role in its short-run fluctuations," he shares.
Martin’s research moves beyond existing studies by using very detailed administrative data from Mexico City to examine how domestic violence responds to temperature, not just day by day, but also neighbourhood by neighbourhood.
Why Mexico City? "My research uses Mexico City as a case study because this city combines several important features: a very large and diverse urban population, an unusually detailed administrative data both on crime and weather, and a serious domestic-violence problem," answers Martin. He adds that, disgracefully, over 40% of women in Mexico City have experienced partner violence during their lifetime.
Moreover, as Martin underlines, Mexico City has a relatively mild climate, due to its high altitude, without too many peaks in heat. "Most existing studies focus on how very hot environments or extreme heat events are linked to more violence," says Martin. "While Mexico City allows me to examine whether temperature matters even in places where ‘extreme heat’ is not the main story – but where small, everyday variations in temperature may still affect stress, comfort, and behaviour inside the home."
How did Martin work to examine this complex topic? To collect data on the domestic violence side, he compared two independent data sources from different reporting channels. The first is judicial reports, which record when and where domestic violence incidents take place across the city. The second is data from a public domestic violence helpline, where women call to seek legal, psychological, or medical support.
Martin explains why using both sources is crucial: "These two reporting channels capture different ways victims reach out for help. Using both allows me to check that the patterns I observe are not simply driven by changes in how people report violence," he points out, adding that this is an essential aspect for a crime that is so heavily underreported.
For detailed temperature data, Martin relies on hourly readings from Mexico’s national meteorological service, gathered from many weather stations spread across the city and the surrounding areas. He then combines these readings to create accurate daily temperature measurements for each neighbourhood - that reflect the weather conditions residents actually experience.
Martin’s key methodological approach is to compare days that are slightly warmer or slightly cooler than others but otherwise very similar. ''In practice, I compare days within the same neighbourhood, within the same season, and on the same day of the week - such as two Wednesdays in the same month that differ mainly in temperature," he clarifies. This is key to Martin’s research, ensuring his findings only reflect the immediate effect of temperature, without being influenced by differences in location, time of year, or the specific day of the week.
"The main result of my research is that domestic violence increases on warmer days, even in a city with a relatively mild climate like Mexico City," he reveals. "When the day is a few degrees warmer than usual for a particular neighbourhood and time of year, incidents of domestic violence rise," he shares.
Martin’s research shows that the connection between temperature variation and domestic violence is not driven by extreme heat, but rather appears with fairly small shifts in daily temperature, with a 1º rise in daily temperature being associated with approximately a 2.7% increase in domestic violence reports.
"The link between temperature variation and domestic violence is remarkably consistent across both types of data I use," confirms the researcher. “Police reports go up on warmer days, and the same pattern appears in calls to the national domestic violence helpline," he states. "Because these two channels represent very different forms of help-seeking, the fact that both move together strongly suggests that the relationship is not simply a reporting artefact but reflects real changes in violence inside the home."
What is also striking in Martin’s findings is that the effect of daily temperature variation on domestic violence is immediate: Cases increase on the same day as the temperature change, rather than showing up with long delays. This suggests that short-run factors, such as changes in daily routines or how much time people spend at home, may play a role.
A core element of Martin’s analysis is also to understand whether temperature affects all households in the same way. "The results show that it does not. The increase in domestic violence on warmer days is much stronger in poorer neighbourhoods than in richer ones," underlines the researcher. In disadvantaged neighbourhoods, the effect is much more intense: each degree of additional temperature is associated with up to a 50% larger relative increase in domestic violence incidents compared to richer areas of the city.
Martin explains that this is due to several reasons. Poorer areas in Mexico City tend to have housing that is smaller, more crowded, and less insulated from heat. In those areas, many households have limited access to cooling, and outdoor space is scarce. This means that even modest temperature increases can create more discomfort and reduce the options people have to cope with heat.
In contrast, residents in wealthier neighbourhoods are better able to buffer temperature changes through larger living spaces, better ventilation, access to cooling, or the ability to modify their routines more easily. As a result, the link between temperature and domestic violence is much weaker in these areas.
This aspect of inequality is a very important element in Martin’s research. "It shows that the risks associated with temperature are not evenly distributed across the city. Even in a mild climate, environmental stress interacts with economic and housing conditions in ways that place the heaviest burden on the most vulnerable households," he points out.
The EUI researcher enriched his work by writing the paper ‘Temperature Variations and Domestic Violence’ and presenting his research in several working groups and seminars at the EUI. The feedback received from EUI colleagues from different disciplines encouraged Martin to think carefully about how domestic violence is reported, how neighbourhood inequalities shape vulnerabilities, and how to explain these dynamics to a broad audience. Moreover, presenting his paper at international conferences and receiving feedback from scholars working on climate, crime, and urban policy also helped him refine both the empirical strategy and the interpretation of the results.
Martin’s findings take on an added significance in the context of the climate crisis. “As average temperatures rise and warm days become more frequent, the kinds of short-term temperature variations I study are likely to happen more often. This suggests that without adequate adaptation, domestic violence risks may intensify, particularly in poorer neighbourhoods that already face higher vulnerability. In this sense, my research underlines that climate policy is not only about managing environmental change, but also about protecting people’s safety and well-being inside their homes.”
The findings from Martin’s research point to several concrete policy directions.
First, they highlight the importance of improving the quality of housing and basic infrastructure in poorer neighbourhoods. "Many of the households most affected by the temperature–violence link live in homes that trap heat and offer very few ways to cope with it. Investments in ventilation, insulation, shading, and access to cooling—along with the creation of safe, accessible public spaces—could reduce the short-run stress that warmer days generate indoors," says Martin.
Second, his results indicate that support services for victims need to be especially responsive on warmer days. "Domestic violence helplines, social workers, and emergency services could plan for predictable increases in demand linked to temperature fluctuations, just as health systems prepare for higher respiratory cases in winter or heat-related illnesses in summer," he suggests.
Finally, Martin’s research findings underline a broader consideration. "Environmental conditions shape vulnerability, and not all communities have the same capacity to adapt. Policies aimed at reducing domestic violence—whether through prevention, victim support, or community programmes—should take these environmental and spatial inequalities into account," he affirms.
By bringing together detailed data, multiple reporting sources, and a focus on inequality, Martin’s research offers a clearer and more nuanced picture of how temperature shapes domestic violence. Ultimately, the results of Martin’s work suggest that "strengthening the ability of households to cope with everyday heat is also a way of reducing the risk of domestic violence inside the home."
Martin Habets is a doctoral researcher at the EUI Department of Economics. Martin’s thesis entitled ‘Environmental Stress, Political Incentives, and Social Norms: Essays on the Determinants of Individual Behavior’ is supervised by EUI Professor Alessandro Tarozzi (supervisor) and former EUI Professor Thomas Crossley (co-supervisor).
Read the working paper ‘Temperature Variations and Domestic Violence’ on SSRN.