The Community Rating System (CRS) was introduced to encourage community-level flood mitigation and increase household-level National Flood Insurance Program (NFIP) participation. It is not clear, however, if and to what extent community participation in the CRS increases household participation in the NFIP and decreases damage claims payments. We employ genetic matching methods and estimate Mundlak-style panel regression models that control for key geospatial, socioeconomic, and time effects to isolate the CRS treatment effect on these outcomes. Results show a positive and significant effect of CRS participation on NFIP participation. CRS effect on damage claims payments is negative but not significant.