The purpose of this study is to understand the factors that motivate local governments’ participation in a national Community Ratings System (CRS) program. The CRS is a voluntary rating system created by FEMA in 1990 through the National Flood Insurance Program. Its purpose is to encourage cities to engage in flood plain management by assigning ratings that correspond to the communities’ level of flood plain management measures implemented. With participation and lower flood risks, community residents in flood plains enjoy discounted premiums (up to 45%) on federally required flood insurance based on their community’s CRS score. To date, over 1,000 communities are participating in the CRS program, achieving a wide range of ratings.  Using national data on historical CRS participation, the 1990 and 2000 Censuses, finance data from the Census of Governments 1992-2007, climate and topographical information from the United States Department of Agriculture, and other data sources, we analyzed the determinants of CRS participation.  Three basic empirical models were estimated: (1) a logit model to explain why some communities opt to participate and others do not; (2) an OLS model to explain the rating class achieved, given that the community participated; and (3) a tobit model that accounts for the clustering of data at a zero rating.  These models are used to test several competing hypotheses that explain why local governments participate as they do.  Some hypotheses follow political economy rationales, especially in communities where politically active residents or higher potential gains to property tax bases explain why some governments pursue CRS excellence.  Other hypotheses are more political, relying on the structure of the local government (e.g., manager vs. mayor systems) or size of local government as explanatory factors.  Extant flood risk itself, past experience with floods, climate, and topography all might also explain participation, especially under a model where a community participates simply out of risk management concerns.  Finally, we examined also alternative explanations based on social capital promoting community participation in programs like this as well as policy diffusion (perhaps via learning or copycatting) between neighboring communities.  The results shine a light on the drivers for public authorities to engage in costly community risk management for flooding.  This is especially important as flood risks and costs continue to rise.  Moreover, understanding how to better promote community scale risk management and public mitigation efforts remains a major policy challenge for natural disaster risks generally.