Despite numerous studies on the relationship between extreme weather events and societal conflicts, more low-scale types of conflict have received rather little attention. Therefore, we so far know little whether and if so, how extreme weather events can trigger anti-government protests. This paper intends to fill this gap by arguing that these are the sorts of conflict where actually an effect is most plausible. Protests require little formal, long-term organization, are associated with low opportunity costs, involve lower personal risk and entail less complicated coordination than large-scale conflicts. In addition, specific societal circumstances should exacerbate the effect of extreme weather events on protests. Extreme weather events should be more likely to trigger protests if different social groups possess unequal political assertiveness. If weaker social groups struggle to obtain compensation through regular political channels for the adverse consequences of extreme weather events, the incentive for staging protests should increase. Since riots and demonstrations require spontaneous mass-scale mobilization, which is more feasible in urban areas, such a protest-inducing effect of extreme weather events should mainly occur in densely populated areas.
This paper puts the argument to empirical scrutiny. Using a spatial multi-level regression analysis based on grid-months as units of observation, the results provide support for the theoretical argument: Grid-months characterized by extreme weather events are indeed more likely to see protests. This effect is even more pronounced in states where weaker social groups possess little political power and in more densely populated regions.