The NFIP’s Community Rating System (CRS) adopted in the early 1990s is an incentive program that recognizes, encourages, and rewards – by the use of flood insurance premium reductions – community and state activities that go beyond the minimum NFIP requirements. With an access to the entire dataset of the CRS from FEMA, over a period of 15 years (1998-2012) we perform longitudinal analysis of active CRS communities and answer the questions such as: How does the distribution of communities look like across all CRS activities? Do the communities improve their CRS classes over time? How long does it take for CRS communities to move up in class? What’s the tenure in the CRS? Do many communities drop their participation after just a few years in a similar way that many residents drop their flood insurance coverage after just three of four years (Michel-Kerjan et al, 2012)? Overall we find that the CRS program works well. It keeps attracting more communities every year, the tenure is very high with 99% of participation communities remaining in the program from year to year. The distribution of scores across all communities has also improved overtime; the average number of activities communities are involved in increase 20% between 1998 and 2012 (from 10 to 12, out of 18 possible ones). There are positive moves within activities as well.