Extreme temperature and intimate partner violence (IPV): Evidence from Sub-Saharan Africa

Abstract: This paper tests the effect of extreme high and low temperatures on women’s experience of intimate partner violence (IPV) through temperature-induced aggression in Sub-Saharan Africa. Related papers find that drought is likely to increase the incidence of IPV. I find a higher prevalence of physical forms of IPV for women who experienced higher maximum temperatures than the historical average. The prevalence of emotional and sexual violence is lower in this context. A test for the correspondence between the days of temperature-and-humidity-based discomfort and severity of violence reveals the same pattern for the three categories of violence. A woman experiencing extreme heat and cold is more likely to report physical, sexual and emotional violence if her partner drinks and if she is employed. These results highlight the distributional effects of climate change and the need for public policy measures, training of law enforcement agents and amendment of laws to reflect them.

Conference

• Third Biennial Conference on Development, Indira Gandhi Institute of Development Research, Dec 16-19 2024

• Agricultural & Applied Economics Association Annual Meeting, Washington D.C., July 23-25 2023. 

Do political motivations affect recovery from natural hazards? Analysis of floods in India 

Abstract: The state plays a central role in managing risk and responding to natural hazards and climate change, as well as enabling private sector involvement in these areas. However, political processes can potentially undermine recovery if they influence the state's relief efforts. This paper examines whether the following political incentives influence governments’ resource allocation in the aftermath of floods in India: targeting loyal supporters, voters at the margin, or staunch opposition. The paper uses fixed effects linear regression models to analyze differences in the recovery of economic activity of constituencies exposed to a flood shock by election outcomes between 2008 and 2013. It uses a regression discontinuity on the vote share of the ruling party candidate to test the causal effect of a flooded constituency's affiliation with the state ruling party on post-flood economic activity. Economic activity is defined as monthly aggregates of the nighttime luminosity of a constituency. The ruling parties in the sample were more likely to prioritize their staunch opposition in post-flood relief and restoration than their loyal supporters. Increasing vote margin of loss for the ruling party candidate corresponded with faster recovery in the first two months after a flood. A consecutive second term of state governments and the proximity of a flood to an upcoming election were associated with faster recovery. With rising temperatures and climate change, the global incidence of floods is expected to intensify. The results of this paper help understand the political factors that may affect a local community’s recovery and inform the mitigation efforts of state and non-state institutions.

 Email for latest draft.