The National Flood Insurance Program (NFIP) was created in 1968 legislation and today is administered by the Federal Emergency Management Agency (FEMA). When the NFIP was authorized, Congress intended for the program to encourage community initiatives in flood risk management, charge insurance premiums consistent with actuarial pricing principles, and also to make premiums affordable to encourage the purchase of flood insurance by owners of flood-prone properties instead of relying on post-disaster flood aid (NRC, 2015a).
Flood insurance can be purchased by individuals from private insurance agents once their community participates in the NFIP and, in so doing, adopts minimum floodplain management ordinances (NRC, 2015a). Flood insurance policies can be bought directly from the federal government through an NFIP Direct Servicing Agent or from a FEMA-identified Write-Your-Own agent.
NFIP risk-based premiums depend on expected future insurance claims; these claims will depend on coverage selected, property characteristics, and the location of the property. In addition, there are charges added to the premium to cover the costs of administering the program and maintaining a financial reserve to cover catastrophic-loss years. However, from its inception, the NFIP premium structure has deviated from strict adherence to actuarial principles to promote the multiple and sometimes conflicting goals of the original authorization (Hayes and Neal, 2011; NRC, 2015a).
The NFIP has been reauthorized many times since 1968, most recently with the Biggert-Waters Flood Insurance Reform Act of 2012 (BW 2012). In this most recent reauthorization, Congress placed a particular emphasis
on the goal of setting premiums following actuarial pricing principles, motivated by a desire to ensure that future revenues were adequate to pay future claims and administrative expenses (NRC, 2015a). BW 2012 was designed to move the NFIP toward risk-based premiums for all flood insurance policies. The result was to be increased premiums for some policyholders that had been paying less than NFIP risk-based premiums, and to possibly increase premiums for all policyholders (NRC, 2015a).
Recognition of the possibility of increased premiums for some policyholders and broader affordability concerns of flood insurance is reflected in Section 100236 of BW 2012 (Appendix C), and in Sections 9 and 16 of the Homeowner Flood Insurance Affordability Act of 2014 (HFIAA 2014) (Appendix D). These sections called on FEMA to propose a draft affordability framework for the NFIP after completing an analysis of possible options for offering “means-tested assistance” to policyholders for whom higher premium rates may not be affordable.
BW 2012 and HFIAA 2014 mandated FEMA to conduct a study. The legislation also called for FEMA to prepare a “draft affordability framework” as described in Section 9 of HFIAA. In developing the affordability framework, FEMA was to address several matters, only one of which is to propose an assistance program. The required content of the framework (paraphrasing the legislation) included
- a plan for offering targeted assistance for ensuring flood insurance affordability among low-income populations;
- programs to ensure communication of the flood risk to property owners and residents in floodplains;
- recognition of the effectiveness of a full range of individual and community actions to mitigate flood risk in NFIP rating tables;
- a report on the effect of increases in premiums on participation in NFIP; and
- a report on the consequences of map updates on affordability of flood insurance.
Section 100236 of BW 2012 also requested the National Academy of Sciences (NAS) to conduct a study that would “compare the costs of a program of risk-based rates and means-tested assistance to the current system of subsidized flood insurance rates and federally funded disaster relief for people without coverage” (P.L. 112-141; 126 Stat. 957). The reference to programs of “risk-based rates” and “current system” refers to comparing the premiums that would have been in place as a result of BW 2012 with
those before BW 2012. The comparison called for in the congressional request was to be for a time when BW 2012 was in full effect. The 2012 legislative requirement for a NAS study was amended by HFIAA 2014, changing the schedule and providing additional resources for conducting the study (NRC, 2015a).
There are some differences between language in the legislation and the scope of work for this report, and Report 1. As noted in Report 1, “when reading this report, it is important to recognize that the language from Section 100236 of BW 2012 as amended in HFIAA 2014 differs from the language in the statement of task. The language in the statement of task was discussed and agreed on by FEMA and the NAS with consideration of the resources available to the NAS and the needs of FEMA” (see Appendix I for the task statement of both reports).
The first report, entitled “Affordability of National Flood Insurance Program Premiums—Report 1,” described policy options and decisions to be made as FEMA proposes an affordability framework. The summary of that report is included as Appendix A. Report 1 included chapters on the following:
- Background and historical aspects of the NFIP. Chapter 2 described the multiple goals for the NFIP, the initial responsibilities of the private sector, the legacy for current premium-setting practices, and the recent reform legislation.
- Flood insurance pricing, policies, and premiums. Chapter 3 reviewed NFIP premium-setting practices against actuarial pricing principles and explained the rationale for past deviations from those principles for the main types of NFIP policies (NFIP risk-based, pre-flood insurance rate map [FIRM] subsidized, grandfathered, Community Rating System discounted, and preferred risk). The text described how BW 2012 and HFIAA 2014 would affect future policy offerings and premium-setting practices.
- The decision to purchase insurance. Congress has long been interested in promoting the purchase of flood insurance and has continued (as evidenced in HFIAA 2014) to be concerned that higher premiums will discourage purchase. Chapter 4 reviewed the literature on the determinants of insurance purchase decisions and found that the effects of price on purchase was uncertain, but that premiums were one factor—however, not the only factor—affecting the decision to purchase flood insurance.
- The spatial distribution of policy types and location of potential affordability challenges. In Chapter 5, the limited policy data available were used to describe the number and distribution of NFIP policies of different types. Data limitations made it impossible to
determine how many polices were grandfathered, a matter of importance for affordability policy analysis.
- Affordability concepts and a framework for assistance program design decisions. Chapter 6 described three ways to measure when a premium might impose a cost burden on a policyholder and then described six design decision questions that policy makers must consider if they were creating an assistance program: who will receive assistance, what assistance will be provided, how will assistance be provided, how much assistance will be provided, who will pay for assistance, and how will assistance be administered? Technical analysis can provide valuable information, but the final answers to these questions require policy judgments.
- Policy alternatives for an affordability strategy, including direct assistance options and actions that would reduce premiums for all policyholders. Chapter 7 discussed specific ways to offer premium payment assistance and to offer premium-reducing mitigation assistance. For example, the chapter discussed changes to eligibility for mitigation grants. Another part of the chapter discussed alternatives to reduce premiums across the board. For example, the chapter discusses the opportunities and challenges of giving credit to mitigation actions other than elevation for reducing future claims and hence premiums.
This second report proposes an analytical platform and describes the data required for FEMA to use when evaluating the possible alternatives that might be included in an affordability framework. The committee’s charge for this report—Report 2—is outlined in Box 1-1.
In developing the task statement, the NAS committee and FEMA were aware of substantial data gaps for the purposes of evaluating options for an affordability framework.
The committee’s Report 1 concluded that the increase in premium costs to pre-FIRM properties from charging NFIP risk-based rates could only be estimated with additional data on structure elevations. There data are missing for pre-FIRM subsidized (PFS) policyholders, although FEMA is in the process of collecting it, and for polices located outside the special flood hazard area. The committee also found that there is a lack of data to identify whether a current policy is grandfathered. Other data gaps include floodplain property owners’ household income, housing expenses (including mortgage obligations), and perhaps other characteristics for evaluating different means-tested assistance program designs (NRC, 2015a).
Lack of data for evaluating NFIP policy options has been recognized as a challenge in the past. In 1999, in response to the National Flood Insurance Reform Act of 1994, FEMA contracted with PricewaterhouseCoopers
Statement of Task
The Federal Insurance and Mitigation Administration is a component of the Department of Homeland Security, Federal Emergency Management Agency (FEMA), which operates the National Flood Insurance Program (NFIP). On March 21, 2014, President Obama signed the Homeowner Flood Insurance Affordability Act (HFIAA) of 2014 into law. This law repeals and modifies certain provisions of the 2012 Biggert-Waters Flood Insurance Reform Act and makes additional program changes to other aspects of the program not covered by that Act. One modification regards a study being conducted by the National Research Council (NRC) of the National Academy of Sciences. HFIAA requires the submission of the Affordability Study by the FEMA Administrator in 18 months from enactment of the Act.
The second report will propose alternative approaches for a national evaluation of affordability program policy options. The second report will include lessons for the design of a national study from a proof-of-concept pilot study. The second report shall discuss
- data issues such as needs, availability, quantity, and quality;
- appropriate analytical methods and related considerations, including models, computing software, and geographic areas to be analyzed;
- a proof-of-concept pilot analysis will be subcontracted as part of the study. This analysis will apply different methods for conducting a flood insurance affordability analysis for a state (North Carolina) in which data on elevations of structures and hydrologic flood hazards are readily available. This analysis will inform the committee’s deliberations and findings regarding the possibilities for a national-level flood insurance affordability study, for which these data on elevations and flood hazards are less readily available; and
- national implications from the proof-of-concept pilot results including, but not limited to, possible impacts on participation rates (the analytical work for the proof-of-concept pilot may be carried out by the NRC directly or using subcontractors as necessary).
(PwC) for a study on the economic effects of charging actuarially based premium rates for pre-FIRM structures. The PwC report (PwC, 1999) tried to fill elevation data gaps by drawing a sample of pre-FIRM properties and then used sample results to impute missing values to the whole population of pre-FIRM properties including first-floor elevation. Because of study time and cost limitations, a sample of 50 communities that included pre-FIRM subsidized policies (a sample from 15,461 NFIP communities considered in the study) was selected. Elevation data were collected on structures in 23 of the 50 communities. It is not clear from the report the extent to which these
samples were representative of the entire population of NFIP communities. No new effort to secure elevation data was initiated and the 16-year-old results for the PwC report continue to be used to estimate the elevation of pre-FIRM properties in 2015.
More recently, the Government Accountability Office (GAO) and Congressional Research Service reported on various aspects of the NFIP. These included the challenges and financial status (King, 2012), improving the administration of the NFIP (GAO, 2011), subsidized properties (GAO, 2013), and strategies for improving the role of the private sector (GAO, 2014a) (see Appendix A for further description of the reports). In 2014, GAO also reported on forgone premiums—the difference between subsidized and full-risk premiums. As of the end of September 2013, there were more than 1.1 million subsidized policies (GAO, 2014b). GAO found that forgone premiums could not be measured, as there was a lack of property elevation data for PFS policyholders. Nonetheless, GAO did estimate how much forgone premiums might be and this ranged between $16 and $25 billion for the period 2002 to 2013 (GAO, 2014b). To do that, they relied upon estimates provided by FEMA (which in part are based on the elevation data reported in the 1999 PwC study).
The reality of limited data and little analytical capacity to quantitatively determine outcomes of different flood insurance affordability policy options is the context for this report. Importantly, the recognition that data gaps would not be easily filled directed the task statement for Report 2 toward a report on analytical process design, data gap identification, and approaches to filling data gaps.
Responses to the questions implied by Congress in BW 2012 and HFIAA 2014 are being developed by FEMA in a context of no existing analytical platform and significant data gaps (see NRC, 2015a; GAO, 2013; and King, 2013). Chapter 2 of this report describes various models with a focus on a microsimulation approach to policy analysis for the NFIP that can be structured and scaled to the available time, cost, and data resources and then enhanced as more resources become available. This analytical approach, considered in the context of questions that need to be answered for an affordability framework, provides a basis for understanding data requirements. Chapter 3 then discusses the available data both in the NFIP policy database and from other sources both in and outside of FEMA. Presently, much of the data needed for simulation are not available; therefore, Chapter 3 also reports on ways to fill those data gaps. Within Chapters 2 and 3, the committee makes reference to a report prepared by the North Carolina Floodplain Mapping Program. This report used the analytical
method of microsimulation to evaluate affordability policy options and, in so doing, served as a proof-of-concept for that method and helped identify data needed to perform affordability policy analyses. Chapter 4 has two main sections. The first includes suggestions for near-term analysis that can be accomplished with existing or modestly expanded resources and data. In the process of preparing its second report, the committee had the opportunity to reflect on the findings of Report 1 and, as a result, developed additional findings (all findings are in boldface type and the numbering of findings reflects chapter number and sequence in the respective chapter) for consideration by FEMA as it prepares its affordability framework for Congress. Therefore, Chapter 4 includes additional findings for some of the topics covered in the different chapters of Report 1. Each chapter presents its findings in the body of the text or at the end of the chapter or both. The report summary includes select finds from Chapters 2, 3, and 4.
Appendix A is the Summary from Report 1. Appendix B is a table of past pertinent reports undertaken by the Government Accountability Office and the Congressional Research Service between 2011 and 2015. Appendix C is Section 100236 from the Biggert-Waters Flood Insurance Reform Act of 2012, and Appendix D is Section 16 of the Homeowner Flood Insurance Affordability Act of 2014. The committee biographical sketches are in Appendix E. Appendix F is a letter sent to the committee requesting evaluation of the specific cost burden measure suggested in HFIAA 2014. Appendix G is a table of data products from the American Community Survey, Appendix H includes tables of data fields found in the NFIP database, and Appendix I includes the task statements for Report 1 and Report 2.
The audience for this report includes FEMA; other relevant federal agencies, such as the U.S. Department of Housing and Urban Development (HUD); Congress and congressional staff; governors of states with flood-prone communities; mayors and citizens in flood-prone communities, especially NFIP policyholders; university faculty and other experts in the fields of natural hazards, flood insurance, and floodplain management; local and state officials with NFIP implementation responsibilities; and private-sector companies involved in flood insurance, flood mapping, and floodplain management.