Disaster Assistance: Information on the Cost-Effectiveness of Hazard Mitigation Projects

Published by the Government Accountability Office on 1999-03-04.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

                    United States General Accounting Office

GAO                 Testimony
                    Before the Subcommittee on VA, HUD, and Independent
                    Agencies, Committee on Appropriations, U.S. Senate

To Be Released
at 9:30 a.m. EST
                    DISASTER ASSISTANCE
March 4, 1999

                    Information on the
                    Cost-Effectiveness of
                    Hazard Mitigation Projects
                    Statement for the Record by
                    Stanley J. Czerwinski, Associate Director,
                    Housing and Community Development Issues,
                    Resources, Community, and Economic
                    Development Division

    Mr. Chairman and Members of the Subcommittee:

    This statement for the record provides our preliminary views on how the
    Federal Emergency Management Agency (FEMA) ensures the
    cost-effectiveness of projects funded under the Hazard Mitigation Grant
    Program. We are conducting this work at the request of this Subcommittee
    and the Chairman of the Subcommittee on Oversight, Investigations, and
    Emergency Management, House Committee on Transportation and

    For a number of years, the Congress has been concerned about the
    increasing costs of federal disaster assistance. One of FEMA’s primary
    approaches for reducing these costs is to promote mitigation measures
    that will reduce future damage within communities—potentially
    decreasing future federal disaster expenditures. However, there are
    concerns that FEMA’s mitigation funding is not targeted to cost-effective
    measures, as mandated by the Robert T. Stafford Disaster Relief and
    Emergency Assistance Act. Our statement is based on previous and
    ongoing work and provides (1) an overview of the increases in disaster
    assistance costs and FEMA’s mitigation programs and (2) our preliminary
    views on the approaches FEMA uses to ensure that funding under the
    Hazard Mitigation Grant Program is targeted to cost-effective mitigation

    In summary:

•   Federal disaster assistance costs billions of dollars annually. For disasters
    that occurred between 1989 and 1993, average annual obligations in FEMA’s
    disaster relief fund totaled $1.6 billion, in 1998 dollars, while average
    annual obligations over the past 5 years (1994 through 1998) have
    increased to $2.5 billion annually in 1998 dollars (even with the exclusion
    of one of FEMA’s costliest disasters—California’s Northridge earthquake).
    The growth in disaster assistance costs in the 1990s has been attributed to
    a number of factors, including a sequence of unusually large and costly
    disasters; an increase in the number of presidential disaster declarations;
    and a gradual expansion in eligibility for assistance. To reduce these costs,
    FEMA is using, among other things, hazard mitigation efforts. These efforts
    promote community involvement in mitigation measures by providing
    grants and training to state and local governments. FEMA’s efforts include
    providing federal flood insurance, converting flood-prone properties to
    open space, mitigating damage to public facilities, reducing earthquake
    risks, and helping mitigate the loss of life and damage from fires.

    Page 1                                                      GAO/T-RCED-99-106
             •   Our ongoing review of FEMA’s efforts to ensure the cost-effective use of
                 federal dollars for hazard mitigation has focused on the Hazard Mitigation
                 Grant Program—one of FEMA’s primary sources of funding for
                 implementing hazard mitigation measures within communities. FEMA uses
                 benefit-cost analysis1—an approach recommended by the Office of
                 Management and Budget—as its primary approach for ensuring that
                 mitigation measures within the Hazard Mitigation Grant Program are
                 cost-effective. However, FEMA also excludes certain types of Hazard
                 Mitigation Grant Program projects from benefit-cost analysis—including
                 projects that fund the removal of certain structures from floodways,
                 research for new building codes, and planning efforts. FEMA officials stress
                 a need for flexibility in assessing these projects, citing the difficulties of
                 quantifying the benefits of some projects and the time needed to gather
                 data to conduct a benefit-cost analysis. However, these exemptions limit
                 the agency’s ability to demonstrate that the funded mitigation measures
                 are cost-effective. Additionally, according to our review of selected
                 benefit-cost analyses in two FEMA regions, officials conducting these
                 analyses were generally knowledgeable and had been trained in how to
                 conduct the analyses. However, they did not always use the best available
                 information in analyzing projects designed to mitigate future damage from
                 flooding events. For example, the officials did not always use flood
                 damage information available from past insurance claims.

                 Following a disaster, at the request of a state governor, the President may
Background       issue a major disaster declaration for the affected areas, thus triggering a
                 range of assistance from federal agencies. The costs of this disaster
                 assistance have grown notably between the late 1970s and 1990s. Between
                 1979 and 1988, FEMA’s obligations in its disaster relief fund exceeded
                 $500 million only in 1 year. In comparison, since 1989, the obligations in
                 the fund have exceeded $1 billion every year except for 1991. The increase
                 in costs is also seen in the number of large, costly disasters. Prior to 1989,
                 only Hurricane Agnes cost the fund in excess of $500 million, while 10
                 disasters have cost over $500 million since 1989. While FEMA has
                 implemented a number of approaches to reduce the costs of disaster
                 assistance—such as consolidating multiple disaster response and recovery
                 functions at individual disaster sites to reduce administrative costs—the
                 agency has made disaster mitigation a primary goal in its efforts to reduce
                 the long-term costs of disasters.

                  Benefit-cost analysis is used to determine how the anticipated dollar savings gained through
                 implementing a project compare with its cost. In order to be considered cost-effective, a project must
                 return more money over its life than it cost.

                 Page 2                                                                          GAO/T-RCED-99-106
FEMA’s September 1997 strategic plan, entitled “Partnership for a Safer
Future,” states that the agency is concentrating its activities on reducing
disaster costs through mitigation because “no other approach is as
effective over the long term.” Mitigation activities are undertaken to
reduce the losses from disasters or prevent such losses from occurring.
The agency’s hazard mitigation efforts include grants and training for state
and local governments; funding for mitigating damage to public facilities;
the purchase and conversion of flood-prone properties to open space;
federal flood insurance; the development of land-use plans and zoning
ordinances to discourage building in hazardous areas; and programs
targeted at reducing the loss of life and property from earthquakes and

However, as we noted in previous testimony,2 quantifying the effects of
mitigation efforts can be difficult. Specifically, determining the extent to
which cost-effective mitigation projects will result in federal dollar savings
is uncertain because the savings depend on the actual incidence of future
disasters and the extent to which the federal government would bear the
resulting losses.

The Stafford Act requires that hazard mitigation measures under the
Hazard Mitigation Grant Program be cost-effective and that they
substantially reduce the risk of future damage, hardship, loss, or suffering.
According to Office of Management and Budget (OMB) guidelines,
contained in OMB Circular A-94, the use of benefit-cost analysis is the
recommended approach for determining cost-effectiveness. FEMA’s
guidance for determining the cost-effectiveness of hazard mitigation
projects3 states that “a key criterion for mitigation projects to be eligible
for funding is that they must be cost-effective” and that “benefit-cost
analysis is used for all cost-effectiveness determinations.”

Benefit-cost analysis is used to assess whether the expected costs of
investing in a hazard mitigation project are justified because the project
will help avoid damages expected from future disasters (the benefits).
FEMA generally conducts the benefit-cost analysis for the projects that

  Disaster Assistance: Information on Federal Disaster Mitigation Efforts (GAO/T-RCED-98-67, Jan. 28,
How to Determine Cost-Effectiveness of Hazard Mitigation Projects, A New Process for Expediting
Application Reviews, Interim Edition, Dec. 1996.

Page 3                                                                         GAO/T-RCED-99-106
                          states submit for approval.4 By conducting a benefit-cost analysis, the
                          analyst determines a benefit-cost ratio—the ratio of the expected benefits
                          divided by the expected costs. If the expected benefits are greater than the
                          expected costs, the ratio is greater than 1.0 and the project is considered
                          cost-effective. If the expected benefits are less than the expected costs,
                          the ratio is less than 1.0 and the project is considered not cost-effective.
                          FEMA’s guidance describes four main elements of a benefit-cost analysis:

                      •   an estimate of damages and losses before mitigation,
                      •   an estimate of damages and losses after mitigation,
                      •   an estimate of the frequency and severity of the hazard causing the
                          damages (such as the risk of flooding), and
                      •   economic factors used in the analysis (a project’s expected life span, for

                          After all of these elements are considered, along with a project’s expected
                          costs, a project’s cost-effectiveness can be determined. However, other
                          factors outside of the benefit-cost analysis can also influence whether a
                          project is accepted for funding, such as the project’s potential impact on
                          environmental conditions.

                          Federal disaster assistance costs have increased in the 1990s for several
Growth in Federal         reasons, including several unusually large and costly disasters, increasing
Disaster Assistance       population and development in hazard-prone areas, increases in the
Costs                     federal share of disaster assistance costs in larger disasters, an upward
                          trend in the annual number of presidential disaster declarations, and an
                          increase in the types of facilities eligible for disaster assistance. Total
                          obligations from FEMA’s Disaster Relief Fund for the 10-year period prior to
                          1989 were $4 billion; since 1989, they have totaled $25 billion.5

Factors Underlying        The large disaster assistance costs in the 1990s have been attributed to a
Increasing Costs          number of factors. Since 1989, the United States has experienced a series
                          of unusually large and costly disasters, including Hurricane Hugo,
                          Hurricane Andrew, the 1993 Midwest floods, and the Northridge
                          earthquake. Hurricane Georges was added to this list in 1998—FEMA is

                           Three states (Florida, North Dakota, and Ohio) typically conduct the benefit-cost analysis for projects
                          from their communities and submit a sheet summarizing the analysis for FEMA’s review. These states
                          have been given additional responsibilities as participants in a pilot program called the “managing
                          state concept.”
                           Since these figures are expressed in nominal dollars, they do not reflect the effects of inflation over
                          the time periods cited.

                          Page 4                                                                             GAO/T-RCED-99-106
                     projecting that it might be the agency’s second costliest disaster ever. The
                     close occurrence of such costly disasters in the United States is
                     unprecedented. Furthermore, increases in population and development,
                     especially in hazard-prone areas, increase the potential losses associated
                     with these disasters. For example, FEMA expects that by 2010 the number
                     of people living in the most hurricane-prone counties (36 million in
                     1995) will double.

                     For several of these large disasters, the federal government has increased
                     its share of the disaster relief costs to provide additional assistance to the
                     states. For example, while the federal share of funding is at least
                     75 percent for assistance to repair or replace disaster-damaged public and
                     nonprofit facilities, the President used his authority to raise the federal
                     share to 90 percent for the Northridge earthquake and to 100 percent for
                     Hurricane Andrew.

                     There has also been an upward trend in the annual number of presidential
                     disaster declarations. From fiscal years 1989 through 1993, the average
                     number of major disaster declarations was 38 per year, while from fiscal
                     years 1994 through 1998, the average number increased to 49.

                     Additionally, over the years, the Congress has generally increased
                     eligibility by expanding the categories of assistance and/or specified
                     persons or organizations eligible to receive assistance. For example, a
                     1988 law expanded the categories of private nonprofit organizations that
                     are eligible for FEMA’s public assistance program.

                     According to a report by the Senate Bipartison Task Force on Funding
                     Disaster Relief,6 federal budgeting procedures for disaster assistance may
                     also have influenced the amounts appropriated for disaster assistance.
                     This is because disaster relief appropriations have often been designated
                     as “emergency” spending, thus excluding them from the strict budget
                     disciplines that apply to other spending. Some views in the report
                     suggested that the assistance provided is more generous than would be the
                     case if it had to compete with other spending priorities.

                     To reduce disaster assistance costs, one of FEMA’s primary approaches has
FEMA’s Hazard        been to emphasize hazard mitigation through various incentives.
Mitigation Efforts   Mitigation consists of taking measures to prevent future losses or to

                       Federal Disaster Assistance, Document No. 104-4, U.S. Senate (Washington, D.C.: U.S. Government
                     Printing Office, 1995).

                     Page 5                                                                        GAO/T-RCED-99-106
                        reduce the losses that might otherwise occur from disasters. For example,
                        floodplain management and building standards required by the National
                        Flood Insurance Program might reduce future costs from flooding. FEMA
                        estimates that the building standards that apply to floodplain structures
                        annually prevent more than $500 million in flood losses.

A Number of Programs    FEMA  funds or otherwise promotes hazard mitigation through a number of
Provide for Hazard      programs. As part of its National Flood Insurance Program, FEMA attempts
Mitigation Assistance   to reduce future flood losses by providing federally backed flood
                        insurance to communities that adopt and enforce floodplain management
                        ordinances that help mitigate the effects of flooding upon new or existing
                        construction. This program also funds a flood mitigation assistance
                        program through the National Flood Mitigation Fund. In 1998, FEMA
                        distributed over $14 million to states and communities to plan and
                        implement measures to reduce future flood damage in homes and other
                        properties that had experienced repeated losses from flooding. Eligible
                        projects under this program include elevating structures, flood-proofing
                        properties, and buying out and converting flood-prone properties to open

                        FEMA also provides grants to states to prevent or reduce the risks of
                        earthquakes by using mitigation measures such as the seismic retrofitting
                        of buildings. The agency also conducts training, public education, and
                        research programs in subjects related to fire protection technologies. The
                        agency’s efforts support the nation’s fire service and emergency medical
                        service communities through such services as the national fire incident
                        reporting system, which collects and analyzes data in order to help
                        mitigate the loss of life and damage from fires.

                        In 1997, FEMA began Project Impact—an initiative based on the premise
                        that consistently building safer and stronger buildings, strengthening
                        existing infrastructures, enforcing building codes, and making proper
                        preparations prior to a disaster would save lives, reduce property damage,
                        and accelerate economic recovery. The initiative intended to build
                        “disaster-resistant communities” through public-private partnerships, and
                        it included a national awareness campaign, the designation of pilot
                        communities showcasing the benefits of disaster mitigation, and an
                        outreach effort to community and business leaders. Project Impact
                        received an appropriation of $25 million in the fiscal year 1999 budget.

                        Page 6                                                    GAO/T-RCED-99-106
                          Under section 406 of the Stafford Act, communities recovering from
                          disasters can use federal funds to mitigate future damage to public
                          facilities that have been damaged. For example, as a damaged building is
                          rebuilt, seismic retrofitting is added to help reduce damages from future
                          earthquakes. Mitigation measures funded under the section 404
                          program—the Hazard Mitigation Grant Program—differ from the 406
                          program in that they can be targeted to either damaged or undamaged
                          facilities. For example, putting storm shutters on the windows of
                          structures is expected to help mitigate wind and rain damage from future
                          hurricanes. Our statement focuses on the measures funded under the
                          Hazard Mitigation Grant Program.

Hazard Mitigation Grant   Under the Hazard Mitigation Grant Program, up to 15 percent of the total
Program                   funds spent on a disaster may be spent specifically on hazard mitigation
                          measures. Subject to certain dollar limits, the act generally allows the
                          funding of up to 75 percent of the cost of hazard mitigation measures
                          within communities that have been affected by a disaster7 (the states or
                          local governments pay the remaining portion of the costs). In fiscal year
                          1998, FEMA approved and obligated over $415 million in Hazard Mitigation
                          Grant Program grants. These grants can be used to protect either public or
                          private property, including the acquisition and relocation of structures
                          from hazard-prone areas. The Stafford Act establishes that the federal
                          contribution is based on measures that “the President has determined are
                          cost-effective and which substantially reduce the risk of future damage,
                          hardship, loss, or suffering in any area affected by a major disaster.” The
                          program funds a range of projects, including purchasing properties in
                          flood-prone areas, adding shutters to windows to prevent future damage
                          from hurricane winds and rains, or rebuilding culverts in drainage ditches
                          to prevent future flooding damage.

                          Historically, hazard mitigation has been considered primarily a
                          responsibility of local and state governments as well as private citizens,
                          since these entities often control the decisions affecting hazard mitigation.
                          For example, building code enforcement and land-use planning are
                          generally under local jurisdictions. As a result, FEMA works with state and
                          local governments to instill a community-based approach to implementing
                          disaster mitigation efforts. Section 409 of the Stafford Act plays a role in
                          developing this approach because it helps to establish the requirement for

                           In an October 10, 1997 regulation, FEMA announced that for disasters declared after April 6, 1997,
                          eligibility for program funding would be statewide rather than limited to the communities affected by
                          the disaster. FEMA was attempting to give the states enhanced flexibility in using the funding for
                          priority projects across the states and to expedite closing out the funding from older disasters.

                          Page 7                                                                         GAO/T-RCED-99-106
                            a comprehensive state hazard mitigation plan that includes an evaluation
                            of a state’s vulnerability to natural hazards. Additionally, as a condition of
                            receiving a Hazard Mitigation Grant Program grant, the state must prepare
                            an administrative plan that establishes its procedures and priorities for
                            identifying and selecting mitigation projects. FEMA, however, has final
                            approval authority for funding these projects. FEMA guidance states that an
                            “ideal” plan would include a statewide mitigation strategy and identify
                            potential hazard mitigation projects that are consistent with the plan.

                            We talked with FEMA staff responsible for approving these plans and
                            reviewed plans from several states. In general, we found that state
                            administrative plans exhibited a broad range of approaches for identifying
                            and selecting mitigation projects. Additionally, a 1996 study8 found that
                            many of the 39 state plans reviewed were “merely intended to qualify the
                            state for post-disaster mitigation grants under section 404 of the Act.” FEMA
                            officials generally agreed with this conclusion. However, several officials
                            noted that the agency has recently initiated changes to improve the states’
                            planning efforts.

                            Our preliminary review found that FEMA’s guidance recommends the use of
FEMA Does Not               benefit-cost analysis as the primary approach for determining a project’s
Always Use                  cost-effectiveness. However, the agency excludes certain categories of
Benefit-Cost Analysis       Hazard Mitigation Grant Program projects from this analysis. These
                            categories include projects that fund the removal of certain structures
to Determine                from floodways, tornado-related measures, research for new building
Cost-Effectiveness          codes, and planning efforts. While FEMA has explained the rationales for
                            these exemptions, certain factors, such as the lack of an analytical basis
and at Times Does           for an exemption on the acquisition of certain floodplain properties, are
Not Use Best                limiting the agency’s ability to demonstrate that these mitigation measures
Available Data              are in fact cost-effective.

Certain Types of Projects   The Stafford Act requires that Hazard Mitigation Grant Program projects
Exempted From               be cost-effective. FEMA’s guidance establishes that benefit-cost analysis is
Benefit-Cost Analysis       the preferred method for making this determination. However, since
                            September 1996, FEMA has exempted the following four categories of
                            Hazard Mitigation Grant Program projects from the use of benefit-cost

                             Edward J. Kaiser and R. Matthew Goebel, Analysis of Content and Quality of State Hazard Mitigation
                            Plans Under Section 409 of the Stafford Act, June 1996.

                            Page 8                                                                        GAO/T-RCED-99-106
                            •   projects involving the purchase of substantially damaged structures in
                                100-year floodplains;
                            •   up to 5 percent of the Hazard Mitigation Grant Program funding for a
                                variety of hazard mitigation measures, such as disaster warning systems or
                                the application of new, unproven mitigation techniques;
                            •   hazard mitigation planning projects for older disasters; and
                            •   an additional 5 percent of the Hazard Mitigation Grant Program funding
                                for tornado-related projects.

                                FEMA’s  general rationale for the exemptions varies, although the agency’s
                                policy guidance establishes that two of the exemptions were made
                                because some mitigation projects were often difficult to evaluate against
                                “traditional quantitative program cost-effectiveness and eligibility
                                criteria.” FEMA officials have explained that the benefits of some projects
                                are difficult to quantify against known project costs and that the time
                                involved in gathering the data on some mitigation projects can be
                                excessive. For example, it is difficult to determine the benefits of
                                establishing an educational program that uses fliers to inform the public
                                about the risks of living in a floodplain because it is hard to predict the
                                resulting changes in public behavior that might result from the fliers.
                                However, without any measurement and subsequent comparison of a
                                project’s expected benefits with its expected costs, it is unclear what
                                criteria the agency is using to determine cost-effectiveness.

Exemption of Projects           Through policy guidance established in September 1996, FEMA exempted
Involving the Purchase of       projects that involved purchasing structures located in floodways and
Substantially Damaged           floodplains—if the cost of restoring the damaged structures equaled or
                                exceeded 50 percent of the structures’ market value and the structures
Structures                      were located in a 100-year floodplain. This particular exemption has come
                                under criticism by FEMA’s Inspector General. In a March 1998 report,9 the
                                Inspector General questioned the exemption’s lack of analytical data
                                supporting the contention that acquisition projects involving substantially
                                damaged properties in the 100-year floodplain were cost-effective. While
                                FEMA officials have begun to retroactively analyze some of the acquisition
                                projects exempted under this policy, the agency is currently unable to
                                provide the analytical data that would support exempting all substantially
                                damaged structures in a 100-year floodplain. FEMA officials explained that
                                they need to conduct a detailed and rigorous analysis of acquisition
                                projects to support the policy. Without this analytical basis, it is difficult

                                 Improvements Are Needed in the Hazard Mitigation Buyout Program, FEMA OIG, Inspection Report
                                I-01-98, March 1998.

                                Page 9                                                                   GAO/T-RCED-99-106
                           for FEMA to demonstrate that the exempted acquisition projects it is
                           funding are cost-effective.

Exemption of Up to 5       In September 1996, FEMA established another policy that exempted
Percent of the Hazard      projects from benefit-cost analysis. Known as the “5 percent Hazard
Mitigation Grant Program   Mitigation Grant Program initiatives,” this policy allowed the states to use
                           up to 5 percent of their Hazard Mitigation Grant Program project funding
Funding for Various        for a variety of hazard mitigation measures. According to FEMA’s policy
Projects                   memo for this exemption, the evaluation of funding for certain mitigation
                           measures, such as hazard warning systems or research for new building
                           codes, required a large amount of time at the state and federal levels,
                           although it was generally recognized that such measures reduced the
                           potential losses from a future disaster. The policy was intended to provide
                           the states with discretion in deciding which mitigation measures they
                           wanted funded, as well as the responsibility for providing the rationale for
                           the cost-effectiveness of the projects selected. FEMA officials explained
                           that the intent of the policy was to spur creativity and avoid the time and
                           expense involved with conducting a benefit-cost analysis.

                           To be eligible, a project type had to be identified in the state’s hazard
                           mitigation plan and reduce or prevent future property damage, injury, or
                           the loss of life. Instead of conducting a benefit-cost analysis, the states
                           were instructed to include a narrative that identified the mitigation
                           benefits and the reasonable expectation that future property damage,
                           injury, or the loss of life would be reduced or prevented. In fact, FEMA’s
                           guidance instructs project applicants to use 5-percent funding if the
                           project was “previously denied because of difficulty in measuring
                           cost-effectiveness.” While FEMA’s guidance instructs the states to identify a
                           project’s benefits, it does not specifically suggest any comparison of the
                           benefits with the project’s costs or competing alternative projects. Without
                           any measurement and subsequent comparison of a project’s expected
                           benefits with its expected costs, the criteria the agency is using to
                           determine cost-effectiveness are unclear. Additionally, by using such a
                           broad determination of a project’s cost-effectiveness, it appears that
                           almost any project could be determined as cost-effective.

Exemption of Hazard        About 1 year later, in October 1997, FEMA announced its third policy
Mitigation Planning        decision, when it exempted hazard mitigation planning projects associated
Projects for Older         with older disasters from benefit-cost analysis. FEMA decided that in the
                           interest of expediting the closeout of disasters that occurred on or after

                           Page 10                                                    GAO/T-RCED-99-106
                           June 10, 1993, the agency would make remaining program funds from
                           these disasters available for hazard mitigation planning purposes.10 States
                           were invited to submit Hazard Mitigation Grant Program applications for
                           funding that would help them develop multi-hazard mitigation plans. The
                           policy memo stated that “funds provided for planning purposes shall be
                           considered a cost-effective measure.”

Exemption of Up to 5       In August 1998, FEMA announced the fourth policy exempting certain
Percent of the Hazard      projects from benefit-cost analysis. FEMA extended its 5-percent set-aside
Mitigation Grant Program   funding by another 5 percent to fund tornado-related projects. The agency
                           noted an increase in tornado activity that it associated with the 1997-98 El
Funding for                Nino weather pattern and suggested that the need for additional funding
Tornado-Related Projects   for warning systems could not be accommodated through existing
                           programs. In essence, the agency increased the 5-percent set-aside policy
                           to a 10-percent set-aside policy, although the additional 5 percent of
                           Hazard Mitigation Grant Program funding was limited to states that had
                           received a presidential disaster declaration for tornadoes. In addition to
                           including a narrative that identified the project’s mitigation benefits and
                           the expectation that future damage or loss of life or injury would be
                           reduced or prevented, the states were required to develop a
                           comprehensive plan for warning their citizens, including a public
                           education component. The policy applied to all disasters with unobligated
                           funds that were declared before fiscal year 1998, as well as all fiscal year
                           1998 and future declarations in which tornadoes or high winds played a
                           role. The policy remains in effect until FEMA adopts proposed regulatory
                           changes stating that warning systems will only be funded from the original
                           5-percent set-aside. FEMA officials expect that the regulatory changes will
                           be made final in mid-March 1999.

Estimating the Number      We are working with FEMA to quantify the number and dollar amount of all
and Dollar Figure of       of the Hazard Mitigation Grant Program measures exempted from
Hazard Mitigation Grant    benefit-cost analysis. However, for a number of reasons, FEMA is unable to
                           readily provide us with this information for all of the exempted projects.
Program Grants Exempted    For example, it is hindered in providing this information because there is
From Benefit-Cost          no data field in the Hazard Mitigation Grant Program database that would
Analysis                   allow the agency to specifically identify the projects that fall under the
                           exemption for acquiring property that has been substantially damaged.

                             When the Hazard Mitigation Grant Program was established, it provided federal matching grants on a
                           cost-share basis of up to 50 percent of a project. Thus, FEMA refers to these mitigation projects as
                           “50/50 planning” projects. With the 1993 amendments to the Stafford Act, the federal cost share was
                           changed from up to 50 percent to up to 75 percent.

                           Page 11                                                                       GAO/T-RCED-99-106
                             Additionally, agency officials have expressed reservations about the
                             accuracy of the data. For these reasons, our preliminary numbers are
                             limited to the 55 hazard mitigation project files we examined for four
                             states (Arkansas, Florida, Louisiana, and Texas) in FEMA regions 4 and 6.

                             These 55 projects represented approximately $20 million in hazard
                             mitigation grant funding, with Florida accounting for 36 projects, or
                             $17.2 million of the amounts reviewed, while the other states accounted
                             for the remaining 19 projects, or approximately $2.8 million in funding. Of
                             the 55, 14 (25 percent), or over $8 million (42 percent) of the funding, were
                             exempted from benefit-cost analysis. One-half of the exempted projects
                             were property acquisitions, while the remaining exempted projects
                             included funding for emergency satellite communications, all-weather
                             radios, emergency alert systems, and a public awareness campaign. The 41
                             remaining projects subjected to benefit-cost analysis included wind
                             retrofits (shutter projects), drainage improvements, seismic retrofits of
                             buildings, and the installation of gas shut-off valves in structures.

Some Benefit-Cost            In the four states we reviewed, the officials conducting the benefit-cost
Analyses Conducted on        analysis were generally knowledgeable about the process and had
Acquisition Projects Do      received training on how to use FEMA’s computerized modules. However,
                             we also found that the officials did not always use the best available data
Not Use the Best Available   for estimating the benefits of projects involving the acquisition of property
Data                         located in floodplains. These data help determine the extent of the
                             expected benefits attributed to a project and significantly influence the
                             accuracy and final outcome of the benefit-cost analysis.

                             For example, in determining flood hazard data—which establishes the
                             probability and severity of a flood event—FEMA’s guidance suggests using
                             the flood insurance rate maps available through the National Flood
                             Insurance Program.11 These maps establish the number of times a flood is
                             expected to occur in a given area (the frequency of future flooding) and
                             the level of the flooding (its severity). The quality of this information can
                             significantly influence the benefit-cost analysis’ outcome because
                             overestimating the frequency or severity of a flood can inflate the
                             estimated benefits attributed to an acquisition project. We found little
                             evidence that information from flood rate maps was used in the
                             benefit-cost analyses we reviewed. Therefore, we are in the process of

                               The flood hazard data needed is actually found in flood insurance reports which accompany the flood
                             insurance rate maps.

                             Page 12                                                                       GAO/T-RCED-99-106
reviewing several of the analyses to determine how the use of information
from the flood rate maps would have affected the analyses’ outcomes.

We also found that the officials conducting the benefit-cost analysis may
not always use the best available data on damage claims from past
flooding. The quality of this information has a significant influence on the
outcome of the benefit-cost analysis because overestimating the extent of
the damage from a previous flood event can inflate the estimated benefits
attributed to an acquisition project. FEMA officials told us that information
on flood claims available from the National Flood Insurance Program was
not always used, suggesting that they simply used information supplied by
project applicants. We also found that the officials conducting the analysis
do not always validate the damage claims information submitted by the
applicants. As a result, the benefit-cost analysis may rely on testimonial
evidence from the applicant—the individual most likely to benefit from the
acquisition project. We are now working with FEMA to determine if the
agency can easily provide damage claims information from the National
Flood Insurance Program to the officials conducting the benefit-cost

We provided a draft of this statement to FEMA to verify its factual content
and modified the statement where appropriate. Our review was initiated in
December 1998, and it is continuing in accordance with generally accepted
government audit standards.

Page 13                                                    GAO/T-RCED-99-106
Page 14   GAO/T-RCED-99-106
Page 15   GAO/T-RCED-99-106
Related GAO Products

              Disaster Assistance: Information on Federal Costs and Approaches for
              Reducing Them (GAO/T-RCED-98-139, Mar. 26, 1998).

              Disaster Assistance: Information on Federal Disaster Mitigation Efforts
              (GAO/T-RCED-98-67, Jan. 28, 1998).

              Disaster Assistance: Information on Expenditures and Proposals to
              Improve Effectiveness and Reduce Future Costs (GAO/T-RCED-95-140, Mar. 16,

              GAO   Work on Disaster Assistance (GAO/RCED-94-293R, Aug. 31, 1994).

(385785)      Page 16                                                     GAO/T-RCED-99-106
Ordering Information

The first copy of each GAO report and testimony is free.
Additional copies are $2 each. Orders should be sent to the
following address, accompanied by a check or money order
made out to the Superintendent of Documents, when
necessary. VISA and MasterCard credit cards are accepted, also.
Orders for 100 or more copies to be mailed to a single address
are discounted 25 percent.

Orders by mail:

U.S. General Accounting Office
P.O. Box 37050
Washington, DC 20013

or visit:

Room 1100
700 4th St. NW (corner of 4th and G Sts. NW)
U.S. General Accounting Office
Washington, DC

Orders may also be placed by calling (202) 512-6000
or by using fax number (202) 512-6061, or TDD (202) 512-2537.

Each day, GAO issues a list of newly available reports and
testimony. To receive facsimile copies of the daily list or any
list from the past 30 days, please call (202) 512-6000 using a
touchtone phone. A recorded menu will provide information on
how to obtain these lists.

For information on how to access GAO reports on the INTERNET,
send an e-mail message with "info" in the body to:


or visit GAO’s World Wide Web Home Page at:


United States                       Bulk Rate
General Accounting Office      Postage & Fees Paid
Washington, D.C. 20548-0001           GAO
                                 Permit No. G100
Official Business
Penalty for Private Use $300

Address Correction Requested