By Julieth Saenz-Molina (Fordham University)
In Bolivia’s tropical lowlands, cold days are rare. Temperatures rarely dip below 21°C, and when they do, it is often because of a surazo, a sudden cold air mass sweeping up from the south, disrupting agriculture, daily routines, and household life. While 21°C is mild by global standards, in Bolivia’s lowlands, it can be destabilizing.
Using self-reported violence data from the Demographic and Health Survey (DHS), matched to high-resolution daily temperature records, my Job Market Paper examines whether extreme temperatures increase women's risk of intimate partner violence. In low-altitude regions, ten unusually cold days over the course of a year increase IPV risk by about 3.6 percentage points — a meaningful rise in a setting where roughly one in four partnered women already reports violence. Meanwhile, in Bolivia’s high-altitude Andean regions, where cool temperatures are the norm and variations are more stable, the same exposure has no statistically significant effect. The difference is not the temperature itself. It is whether people are adapted to it.
These findings are consistent with the concept of de-adaptation — the idea that populations are most vulnerable not to globally extreme temperatures, but to temperatures that are infrequently experienced locally.
De-Adaptation and Social Vulnerability
Climate discussions often assume absolute temperature thresholds of heat and cold. But social harm does not arise from temperature in isolation. It arises from temperature relative to what communities are built to handle. While a cold day in Minnesota is ordinary, a cold day in tropical Bolivia can disrupt crops, strain household income, and alter daily interactions. The actual temperature matters less than whether homes, labor markets, and social norms are prepared to withstand them.
As temperatures rise, economists often discuss adaptation as a long-run response to hotter temperatures, such as increased use of air conditioning and adjustments to agricultural practices. But climate change does not only shift averages — it also increases variability. As shocks become more frequent and intense, communities will increasingly face conditions outside their historical range, a process that can generate what researchers call harmful de-adaptation.
Bolivia’s lowlands demonstrate what harmful de-adaptation can look like. In these communities, cold spells are rare and unexpected. My findings show that exposure to them increases women’s risk of intimate partner violence not because cold intrinsically causes aggression, but because cold disrupts systems that are not designed for it.
When the Uncommon Arrives
Figure 1 presents the estimated effect of each temperature range on IPV incidence in low and high altitude areas, relative to days with normal temperatures. The pattern is telling. In the lowlands, violence rises at the extremes, not gradually. In the highlands, estimates are small and statistically indistinguishable from zero across the entire distribution. The same country, the same year, but two completely different responses.
Figure 1. Estimated effects of temperature on IPV incidence by altitude
Cold days drive the larger effect — consistent with de-adaptation. But the mechanisms differ by shock type, and so do the populations most at risk.
What explains the lowland pattern? The evidence points to two distinct channels depending on the type of shock. During cold spells, male alcohol consumption rises and labor disruptions signal household income instability, particularly in rural and indigenous communities. Heat tells a different story. In urban areas, extremely hot days reduce women's labor force participation and lower their earnings relative to their partners. That shift in economic position is associated with more severe violence. The mechanism is not alcohol. It is bargaining power.
What This Means for Climate Policy
Most climate damage estimates focus on average temperature increases and assume vulnerability is uniform across populations. My findings suggest this approach misses localized risks. In Bolivia, a national-level analysis shows no significant effect of temperature on IPV. The impacts in the lowlands are offset by the absence of effects in the highlands. The national averages conceal local vulnerability. They flatten heterogeneity into a single coefficient and obscure the harm concentrated in the lowlands.
This has a direct implication for how we design climate adaptation policy. Strategies built around absolute temperature thresholds — assuming that harm rises monotonically with heat — will fail to anticipate cold-shock risks in tropical communities. And as climate change increases variability and shifts baselines, more communities will face unfamiliar conditions. The danger will not always come from record-breaking heat. It may come from moderate temperatures that fall outside what local systems are prepared for.
Protecting women from climate-driven violence requires locally tailored responses: income stabilization during cold spells, support for women's economic participation during heat shocks, and recognition that climate resilience and gender-based violence prevention are not separate policy agendas.
The Bigger Picture
Climate change is reshaping what counts as normal. As that baseline shifts, the social consequences will follow, and they will not be distributed equally. Bolivia's lowlands offer an early warning. The question is whether climate policy is designed to hear it.
About the Author
Julieth Saenz-Molina is a PhD candidate in Economics at Fordham University.
Her research focuses on the intersection of development and environmental economics, examining the socioeconomic impacts of climate variability. To learn more about her work, visit her website: https://www.juliethsaenzmolina.com/
Social Media Handles:
