In the United States, recent decades have seen increases in the economic damages related to floods. Experts have provided three explanations for this trend: (1) climate has been changing such that more or different types of precipitation events have led to more damaging floods, (2) more people are moving to flood prone areas, leading to increased vulnerability, and (3) flood policies have increased exposure to floods (e.g., through subsidies) and/or led to more intense floods (e.g., via channel alterations). While it is likely that climate, demographics, and policy all play some role underlying the damage record, there have been no studies that seek to understand the relative contribution of each to the nation's growing flood damages. This study addresses the relationships of precipitation and damaging floods over the period 1932-1996. Specifically, the paper evaluates and interprets the relationships of ten different measures of precipitation with the flood loss record, both measured at national (1932-1996) and regional (1983-1996) levels. The paper finds that many (but not all) of the different measures of precipitation are significantly related with damages, but that they explain only 10%-20% of the variance in the damage record. The unexplained variance must be explained by variables other than precipitation, including societal (e.g., demographics, policy, etc.) and other climatological (e.g., timing, location, etc.) factors. The paper presents these findings and their significance for the orientation of climatological research focused on understanding hydrological extremes that matter most to society as well as to the policy community seeking to understand the growth in flood losses over recent decades