Not much time has passed since our last blog post on Risk Rating 2.0 but a lot has changed! As the rollout inches closer, new information continues to trickle out about the program. We’re documenting some of our learnings here with a big disclaimer: we are not experts in flood insurance, just stakeholders who have kept a close eye on new developments.
On April 1, 2021 FEMA published a press release titled FEMA Updates Its Flood Insurance Rating Methodology to Deliver More Equitable Pricing. One of the most important pieces of information within the press release was that the rollout of Risk Rating 2.0 will be split into two phases. FEMA cited far reaching economic impacts of COVID-19 as one of the main reasons why and a phased approach has been adopted. The phases are as follows:
In short, those that will face premium decreases under Risk Rating 2.0 will transition to the lower rate immediately at the first renewal of their policy (after October 1, 2021), while premium increases will transition gradually and within the existing statutory limits until the full-risk rate for the property is reached. As stated in previous blog posts this rate is capped at an 18% increase per year. With this new announcement more details have been provided to what will be changing and staying the same as Risk Rating 2.0 is rolled out.
Alongside state-level information about the impacts of Risk Rating 2.0’s changes, FEMA published a new methodology document outlining how premiums will be calculated with Risk Rating 2.0 on April 16th. The document, as well as the corresponding Premium Calculation Worksheet and Appendix, shed some light on how Milliman has developed the rating methodology for Risk Rating 2.0. For those of you who learn best by examining examples, pages 21-23 of the methodology document outline a few rating case studies.
Since its announcement, stakeholders affected by Risk Rating 2.0 have been keen to learn more about the data used in the modeling. It looks like the rating engine will utilize a significant amount of publicly available data, as well as proprietary information/models from vendors like CoreLogic, KatRisk, and AIR. We’re keeping a keen eye on how the use of proprietary data will affect rating transparency going forward. The GIS layers utilized in the methodology are listed as:
We’ve heard from some of our partners that there is some concern over the use of ⅓ Arc-Second DEMs, which have ~10m resolution. In uniquely flood-prone areas of the country, localized conditions can change significantly over small areas. Relatedly, it looks like several of the Correlated Market Basket Fields will utilize distribution by state and year built to estimate values where none are available. These fields include Year Built (for which only distribution by state will be used for estimation), # of Stories, Construction, and Foundation Type. Elevation, BFE, Foundation Type, and Year Built are the factors used to derive First Floor Height in the modeling – an important variable that speaks to exposure. A key feature of our software extracts and mobilizes Lowest Floor Elevations from Elevation Certificates for use by floodplain managers in planning and resident outreach. Because of this, we’ve been keeping a close eye on conversations about the future of ECs and how LFEs factor into rating. The methodology seems to indicate that relatively low resolution data will be used to estimate built conditions. If this is true, it’s likely that Elevation Certificates and ground-truthed data will still play an important role in filling in gaps for rating as well as in advocating for individual premium changes.
According to FEMA, the methodology will give way to the creation of more equitable premiums. The issue of affordability has raised questions and ongoing conversations with Congress. According to Shana Udvardy at the Union of Concerned Scientists, if affordability concerns go unaddressed “the department (FEMA) risks exacerbating the racial and socioeconomic disparities that already exist within flood-prone areas.”
Based on available data, FEMA states that 23% of policyholders will see immediate premium decreases and 77% can expect their premium to stay the same or go up. The data, as explained in the next section, varies greatly by state and it has the possibility of varying greatly between communities and cities in the same state. These rate changes are not as drastic as some expected. Additionally, FEMA has made it clear that Community Rating System discounts will remain. This has huge implications for the affordability of flood insurance in the U.S., particularly in coastal communities that have invested heavily in the program. As Special Flood Hazard Areas will no longer be used to rate premiums, CRS discounts will also be applied to all policies in a community.
Along with the above information, FEMA published a guide on how rates will change across the country. Vale Penguin has also put together some good figures with information about all states exported into one table. Below we’ve highlighted some of the states that will be most affected by the changes in insurance rates.
In terms of absolute numbers, Florida appears to have the most households facing big rate hikes. However, Florida is also the state with the most policies (>2x the amount of policyholders than the next state). Around 73,000 policyholders (4% of the state's total) will see rates increase more than $20/month.
Texas is the state with the second most policies in the country behind Florida. With 86% of these properties seeing an increase with Risk Rating 2.0, Texas will also have the second highest rate only behind Hawaii (which will see 87% of its policies increase in cost).
Only behind Washington D.C. and Alaska (both with less than 2500 policies), 61% of policyholders in Maryland will see their rates fall, while only 1% will see the top category of rate hikes.
All of the New England states except for Rhode Island will see increases larger than the national average on the highest rates (greater than $20 a month). For example, almost 1 in 10 policyholders (9%) in both Connecticut and Maine will see the highest rate increases. This number drops to 8%, 7%, and 6% in Vermont, New Hampshire and Massachusetts, respectively.
FEMA is still working through some of the assumptions that impact areas protected by levees. Finer-grain data hasn’t been made public and it is possible that some municipalities will see significant decreases or increases in insurance rates. The agency is working on a public-facing interface for policyholders that might be available later this summer, which could shed additional light on how Risk Rating 2.0’s methodology will manifest locally. In the meantime, we’ll continue to stay apprised of updates!
This blog post was written in collaboration with our research analyst, Kyle Sweeney. If you have any questions about this article or if you'd like to learn more about what we do please reach out to us at firstname.lastname@example.org.