Natural Disasters, Social Protection, and Risk Perceptions

Natural disasters give rise to loss and damage and may affect subjective expectations about the prevalence and severity of future disasters. These expectations might then in turn shape individuals’ investment behaviors, potentially affecting their incomes in subsequent years. As part of an emerging literature on endogenous preferences, economists have begun studying the consequences that exposure to natural disasters have on risk attitudes, perceptions, and behavior. We add to this field by studying the impact of being struck by the December 2012 Cyclone Evan on Fijian households’ risk attitudes and subjective expectations about the likelihood and severity of natural disasters over the next 20 years. The randomness of the cyclone’s path allows us to estimate the causal effects of exposure on both risk attitudes and risk perceptions. Our results show that being struck by an extreme event substantially changes individuals’ risk perceptions as well as their beliefs about the frequency and magnitude of future shocks. However, we find sharply distinct results for the two ethnicities in our sample, indigenous Fijians and Indo-Fijians, the former of whom are noted for their “collectivist” social organization and the latter of whom are largely self-reliant: the impact of the natural disaster aligns with previous results in the literature on risk attitudes and risk perceptions for Indo-Fijians, whereas they have little to no impact on those same measures for indigenous Fijians. Anchored in the institutional context of Fiji, this result is sensible and has implications for how social protection can mitigate some of the negative impacts of natural shocks on households’ expectations. To provide welfare implications for our results, we compare households’ risk perceptions to climate and hydrological models of future disaster risk, and find that both ethnic groups over-infer the risk of future disasters relative to the model predictions. If such distorted beliefs encourage over-investment in preventative measures at the cost of other productive investments, these biases could have negative welfare impacts. Understanding belief biases and how they vary across social contexts may thus help decision makers design policy instruments to reduce such inefficiencies, particularly in the face of climate change.