Conference Agenda

Frequent occurrences of natural disasters and disaster-related deaths affect government responsiveness to disasters. This study examines how disaster mortality disaggregated by gender and upcoming state elections affected per capita calamity relief spending for Indian states from 1972 to 2019. The results suggest that calamity relief spending increased significantly when more women were killed than men due to disasters, particularly nearer state elections. In addition, the findings highlight that intense political competition is positively associated with relief spending. Increasing male and female electoral turnout also helped improve calamity relief expenditure. This suggests that the government is more responsive to a rise in female disaster mortality via an increase in per capita calamity relief spending in election or pre-election years where the electoral participation of citizens is greater. Our results are robust in different econometric specifications.

The economic costs incurred by extreme weather events are substantial and increasing. In this study, we demonstrate how community banks – a type of financial institution with strong local ties and customer relationships – mitigate these costs. We use an event study model to demonstrate that US counties with higher community bank market shares experience fewer employment losses through extreme weather events. We then use bank-level analyses to demonstrate the mechanism – the small business credit supply. Community banks maintain their lending following extreme weather events, while other banks reduce it. These findings provide novel evidence on how local financial institutions strengthen economic resilience through extreme weather events. As policymakers develop strategies to mitigate the effects of extreme weather events, local fi-nance may be a solution.

The impacts of climate change are already shaping people’s lives today: Natural disasters are increasing in frequency and severity. Innovation is often hailed as the only path to sustainable prosperity. In this article, we study how inventors respond to natural disasters. We combine patent data with the geolocations of European inventors and spatially match these to detailed information on local natural disasters for the years 1994-2014. We exploit this quasi-experimental variation of the disasters to estimate event study designs. Compared to the sample average, we find that in affected areas exposure to natural disasters leads to a persistent increase in green patents of roughly 20%. The strongest disasters lead to an increase of about 50%. We propose in a theoretical model and provide empirical evidence that inventors respond to natural disasters by updating beliefs on future consumer demand for green goods. This effect is stronger in product markets with fierce competition, as green innovation is a way of product differentiation to escape competition. Overall, the effect is primarily driven by increased innovation in mitigation technologies aiming to reduce emissions, thus by the technologies combating the root cause of climate change. We find that disasters have no effect on non-green innovation. Our results highlight an important inefficiency in how innovation responds to climate change. We show that purely local responses to natural disasters yield suboptimal research levels from a welfare perspective. There is an opportunity for policy to act, by propagating the information carried by the disasters beyond locally affected inventors.

Inadequate adaptive responses to flood risk are often attributed to behavioral constraints, such as low awareness, underestimation of risk, overly optimistic beliefs, and reliance on government compensation. Public policies aimed at disseminating current flood risk information could potentially address these distortions.

I use an online survey experiment targeting residents of flood-prone area to assess whether free access to information affects perceived risk and insurance demand. Survey respondents are offered individualized information about the flood risk profile at their address. Moreover, I study the impact of making specific information more visible. I find that survey respondents tend to be “confidently incorrect” about their risk category according to publicly available flood maps. Increased visibility of some, but not all, information affects reading behavior and demand for insurance. However, it does not affect beliefs related to flood risk, damages, and government compensation. I find suggestive evidence of backlash to information among residents of high risk area and individuals underestimating their flood probability. These results underscore that policies aimed at disseminating information on environmental risk can influence households’ adaptation decisions, but not necessarily through the expected channels.