WASHINGTON – U.S. Senators Bill Cassidy, M.D. (R-LA), Cindy Hyde-Smith (R-MS) and Roger Wicker (R-MS) are seeking to ensure the U.S. Government Accountability Office (GAO) considers key facts and data in any legitimate assessment of the methodology used to develop the National Flood Insurance Program’s Risk Rating 2.0 (RR2.0).
Addressed to GAO Comptroller General Gene L. Dodaro, the lawmakers’ letter criticizes the Federal Emergency Management Agency (FEMA) for its continued lack of transparency in developing and implementing RR2.0 and outlines specific questions that should be answered to produce a thorough study of the program.
“A third-party study of Risk Rating 2.0 may be useful if appropriately scoped, but it cannot be a substitute for FEMA making the method totally transparent and implementing it under the rulemaking process of the Administrative Procedure Act,” wrote the senators.
“We request that the following be included as one of the major questions to be examined in the GAO study: ‘To what extent is Risk Rating 2.0 actuarially sound and transparent?’ Our concern lies in that the study as designed might resolve the question of ‘actuarial soundness’ into an examination of whether Risk Rating 2.0 is ‘doing what actuaries do,’” continued the senators.
The senators also express ongoing and growing concerns about the higher premiums levied on policyholders will result in an estimated 20 percent dropping their NFIP coverage. They also question the fiscal soundness of FEMA plans to address the affordability issue.
“A complete study of Risk Rating 2.0 needs to address these discrepancies between affordability and fiscal balance. We would prefer to have enough transparency in this program to face this reality now in order to adopt flexibility with respect to actuarial niceties, and design a program that balances affordability with other objectives,” concluded the senator.
The letter to Dodaro was accompanied by a related July 2022 letter signed by seven senators to FEMA regarding the methodology used in developing and implementing RR2.0 and its implications on reauthorizing the troubled federal flood control program.
Read the full letter here or below.
Dear Comptroller Dodaro:
It is our understanding that you intend to release a Government Accountability Office (GAO) study this year on the National Flood Insurance Program’s Risk Rating 2.0 methodology.
While we would welcome such an evaluation, we would like to note that our immediate concern with Risk Rating 2.0 is that the method and its supporting data have not been made transparent. The Federal Emergency Management Agency (FEMA) has been and continues to be reluctant to share the important details of its computations. A third-party study of Risk Rating 2.0 may be useful if appropriately scoped, but it cannot be a substitute for FEMA making the method totally transparent and implementing it under the rulemaking process of the Administrative Procedure Act. We are enclosing for your information and use a letter to FEMA from our Senate colleagues that expressed these issues and other concerns about Risk Rating 2.0.
We request that the following be included as one of the major questions to be examined in the GAO study: “To what extent is Risk Rating 2.0 actuarially sound and transparent?” Our concern lies in that the study as designed might resolve the question of “actuarial soundness” into an examination of whether Risk Rating 2.0 is “doing what actuaries do.” That concern is further heightened by your intent to assess FEMA’s Risk Rating 2.0 models and methodologies against actuarial standards. In our view, the outcome is likely to be answered in the affirmative given that actuaries designed the method, but the answer will provide little insight in our discussions on what NFIP policies should be.
We are also deeply concerned by FEMA’s report to the U.S. Treasury (Criswell A. D., 2021) that Risk Rating 2.0 will result in attrition of one in five policyholders from the flood insurance program over the next decade. FEMA’s projection raises the dire specter of a million or more homeowners sentenced to living uninsured in their homes. Policyholders’ home values will be devalued and unsaleable, and they will essentially be sentenced to live in their uninsured homes with little help or chance of relief. FEMA has yet to disclose how it will meet this challenge which it has identified as resulting from the implementation of Risk Rating 2.0.
The era of unaffordable flood insurance ushered in by Risk Rating 2.0 led to a new FEMA proposal to provide a means-tested insurance subsidy program to help moderate-to-low-income homeowners afford actuarial premiums. Of course, such subsidized premiums have historically been financed to a large extent within the flood insurance pool itself with contributions to the deficit from the borrowings of the fund. The FEMA proposed subsidy program would depart from actuarial soundness and sacrifice fiscal soundness to rescue Risk Rating 2.0. If the subsidies must be financed within the insurance pool, then NFIP will move a substantial distance back toward the cross-subsidized premiums that FEMA declared Risk Rating 2.0 would deliver us from – all without any clarity that we are better off than with the legacy premium setting method.
A complete study of Risk Rating 2.0 needs to address these discrepancies between affordability and fiscal balance. We would prefer to have enough transparency in this program to face this reality now in order to adopt flexibility with respect to actuarial niceties, and design a program that balances affordability with other objectives.
Beyond these conceptual concerns that we believe your study should examine, specific concerns that have been raised by many in the industry are listed below. We encourage your office to examine these requests and questions and answer them in your report.
- Report the error bands (confidence intervals) from the generalized linear model results for all the rating factors by region. Based on these error bands, evaluate the usefulness of the regressions for estimating flood insurance premiums with an accurate underlying estimate of expected annual damages.
- Evaluate the soundness of estimating rating factors on a regional, as opposed to a flood plain, basis together with discussion of the errors and biases that such an approach may introduce. FEMA reports estimating rating factors based on geographic data that spans Texas to North Carolina rather than on more granular considerations of floodplain relationships.
- Collect a significant sample of property-specific average annual damages using traditional hydrology/hydraulics models from U.S. Army Corps of Engineers (USACE) studies (adjusted for price levels) and compare those results with the same property-specific average annual damage estimates calculated within Risk Rating 2.0.
- Evaluate application of transparent and peer reviewed traditional methods (USACE, Hazard Mitigation) for computing property-specific average annual damages to compute actuarial premiums as compared with the opaque Risk Rating 2.0 methods.
- Evaluate the economic impact analysis in FEMA’s most recent NFIP programmatic environmental impact statement, especially the estimates of impacts on property values. Are these estimates economically sound and do they accurately capture the impacts of eliminating subsidized premiums on property values? Do these estimates accurately depict the impact of the entire Risk Rating 2.0 approach (as opposed to merely eliminating subsidies) on property values?
- What are the immediate and long-term effects of eliminating grandfathered premiums on home values? What impact would this have on local property tax collections by municipalities?
- What is GAO’s estimate of the expected attrition from the program from increasing cost on policyholders who do not have a mandated purchase requirement or others who simply cannot afford to pay the proposed rates?
- Does FEMA have the clear Congressional authority to make the draconian premium increases in Risk Rating 2.0?
- Should FEMA implement Risk Rating 2.0 through the Administrative Procedure Act?
- Should FEMA conduct an independent peer review of all models and methods in Risk Rating 2.0 as required by the Office of Management and Budget guidance for the Information Quality Act?
Thank you very much for your attention to this matter.