Weather Shocks, House Prices, and Population: Role of Expectation Revisions

This paper studies how expectations of future flood risk in communities within the U.S. change once they experience a flood. Focusing on small events, I examine if population and housing values change in a way consistent with higher flood risk. To do this, I compile a new measure of insured and uninsured losses for 4,147 locations and identify small flood events at places with different flood history. I show that flood history determines the extent to which events are anticipated and insured. Flood surprises in communities with high pre-flood growth reduce population and do not affect house prices. In the rest they cause reduction in prices and no population effect