UC San Diego

This dissertation contains three essays exploring how economic agents and markets respond to weather events and natural disasters.

Chapter 1 examines whether short-run weather fluctuations affect the decision to go solar by residential customers in California. The results show that residential customers whose sign-ups are followed by less sunny weather are more likely to cancel their contracts. This behavior is inconsistent with rationality, as a solar PV system is a long-term investment whose return is not affected by short-run weather fluctuations. I analyze this behavior further using a theoretical model of projection bias combined with engineering models of solar power production and energy demand.

Chapter 2 studies housing market responses to hurricanes using detailed data on Florida housing markets during 2000-2016. We identify the causal effects of hurricanes in a difference-in-differences framework exploiting the exogeneity of hurricane path and timing. The results show that the housing market is dominated by a negative supply shock lasting up to three years. While these effects are transitory, we show that they are associated with the arrival of home buyers whose income is higher, holding fixed the quality of homes transacted. This implies a lasting shift in the local economic profile towards higher income, along with potential gentrification.

Chapter 3 investigates whether and how voters’ preferences on environmental policy are reflected in legislative elections. People’s belief in climate change is known to be affected by experiencing extreme weather events. We study the electoral responses to these randomly occurring events in House of Representative elections. To focus on the channel of environmental preference, we estimate differential responses based on the environmental ideology of the incumbent congressperson conditional on party membership. We find evidence that climate-related extreme weather events motivate political giving to challengers of anti-environment incumbents, increase the overall competitiveness in these races, and lead to a lower probability of re-election for these incumbents.