This dissertation studies three aspects of the economics of climate change: how rising sea levels will affect coastal homeowners in Florida; how changes in weather will affect the prevalence of crime in the United States; and why skepticism about climate change is so common among the general public. Chapter 1 uses housing market data to estimate the welfare costs of shoreline loss along coastal beaches in Florida. I develop a structural housing market model and use it to provide a welfare interpretation for the coefficients from a new “discontinuity matching” hedonic research design. Using housing sales data, beach width surveys, and historical beach nourishment records, I then estimate Florida homeowners’ willingness to pay for an extra foot of sand. I find that changes in beach width have little impact on housing prices, except possibly at very eroded beaches. Chapter 2 estimates the impact of climate change on the prevalence of criminal activity in the United States. The analysis is based on monthly crime and weather data for 2,972 U.S. counties from 1960 to 2009. The results show that temperature has a strong positive effect on criminal behavior, and that between 2010 and 2099, climate change will cause an additional 35,000 murders, 216,000 cases of rape, 1.6 million aggravated assaults, 2.4 million simple assaults, 409,000 robberies, 3.1 million burglaries, 3.8 million cases of larceny, and 1.4 million cases of vehicle theft. The social cost of these climate-related crimes is between 20 and 68 billion dollars. Chapter 3 develops a model of rational skepticism about policy-relevant scientific questions. Many policy debates have three features: first, individuals initially disagree about some scientific question; second, new evidence about the question becomes available; and third, the evidence may be systematically biased. Under these conditions, Bayesian disagreements persist even in the face of an infinite quantity of new evidence. Furthermore, Bayesian updating based on the new evidence produces “skeptics”, in the sense that individuals whose prior beliefs conflict most with the observable evidence end up with the most extreme posterior beliefs about the degree of bias. These results provide insight into the phenomenon of climate skepticism.