Authors: Andrew Linke*, University of Utah, Jaroslav Tir, University of Colorado Boulder, Frank Witmer*, University of Alaska Anchorage, John O'Loughlin, University of Colorado Boulder, J. Terrence McCabe, University of Colorado Boulder
Topics: Political Geography, Human-Environment Geography
Keywords: Violent conflict, climate change, floods, Kenya
Session Type: Paper
Presentation File: No File Uploaded
This project leverages the 2018 Kenya floods as a natural experiment to gain insights about the potential link between environmental stress and violence. The floods represent an exogenous environmental shock and provide us with an opportunity to gauge how the respondents’ attitudes about and experiences of violence may have been affected by the flood. Specifically, (1) comparing responses across two survey waves, one before (2014) and the other after (2018) the floods, we are able to determine if the exposure to flooding resulted in a greater likelihood of expressing support for violence and in increased chances that a respondent reports witnessing or being a victim of violence. (2) Exploiting variation across the 2018 survey with respect to flood exposure, we are also able to determine whether flood-affected individuals are more likely to support violence and report being witnesses or victims of violence, compared to individuals not affected by the flood. Across both sets of analyses, we find that the support for violence attitudes remain relatively unchanged. But we find that the exposure to floods leads to significant increases in reports of being a witness to or a victim of violence. The latter findings provide some of the first systematic and direct pieces of evidence that climate change-related environmental stress increase the risk of violence.