The first chapter of the dissertation examines the learning process that economic agents use to update their expectation of an uncertain and infrequently observed event. The standard Bayesian updating model is restrictive in that it reflects the strong neo-classical assumption that economic agents efficiently incorporate new information with all available information when updating beliefs. I consider the case of flooding and estimate the effect of first-hand experience on flood insurance take-up. I compile a new nation-wide panel dataset of large regional floods and flood insurance policies in the US. First, I show that flood insurance take-up in flooded communities increases by 9% after a flood and then steadily declines, fully dissipating after 9 years. Floods do not affect take-up in geographically neighboring non-flooded communities unless these communities are in the same media market. The take-up rate in non-flooded communities that share a media market with a flooded community is one-third as large as in flooded communities. I interpret this evidence using the standard Beta-Bernoulli Bayesian learning model and a Beta-Bernoulli model that includes a forgetting/first-hand experience parameter. I find that the standard Bayesian model can not explain both the spike in insurance in the year of a flood and the decay rate of this effect on insurance take-up in the years after the flood. I conclude that the evidence is most consistent with a Bayesian model augmented with a forgetting/first-hand experience parameter.

The second chapter of my dissertation examines the causal link between localized exposure to hazardous waste pollutants from motor vehicle exhaust and adverse human health outcomes for newborns. I explore whether an exogenous event--the 1994 Northridge Earthquake--can be used as a quasi-experiment to test how birth outcomes change from a sudden and unexpected increase in pollution. The Northridge Earthquake closed down portions of four busy highways in Los Angeles, CA for periods of 1-6 months. The highway traffic was diverted onto secondary roads that previous to the earthquake had a much lower traffic volume. The paper focuses on two health outcomes for newborns: birth weight and gestation period. Infants born preterm or with low birth weight are less likely to survive infancy, more likely to suffer from childhood illness, and have lower future earnings. Overall the results of this study are inconclusive due to the relatively small number of new births included in the sample design. However, the results do suggest that a mother's race, age, and level of education are more important than proximity to a highway. Being a minority race, a teenage mother, or not having any college education are correlated with lower birth weight. The size of these correlations are approximately an order of magnitude larger than the point estimates for the effect of living in close proximity to a road with heavy traffic.

The third chapter of the dissertation uses the housing market to develop estimates of the local welfare impacts of Superfund sponsored clean-ups of hazardous waste sites. We show that if consumers value the clean-ups, then the hedonic model predicts that they will lead to increases in local housing prices and new home construction, as well as the migration of individuals that place a high value on environmental quality to the areas near the improved sites. We compare housing market outcomes in the areas surrounding the first 400 hazardous waste sites chosen for Superfund clean-ups to the areas surrounding the 290 sites that narrowly missed qualifying for these clean-ups. We find that Superfund clean-ups are associated with economically small and statistically indistinguishable from zero local changes in residential property values, property rental rates, housing supply, total population, and the types of individuals living near the sites. These findings are robust to a series of specification checks, including the application of a regression discontinuity design based on knowledge of the selection rule. Overall, the preferred estimates suggest that the local benefits of Superfund clean-ups are small and appear to be substantially lower than the $43 million mean cost of Superfund clean-ups.