This is the accessible text file for GAO report number GAO-08-437
entitled 'National Flood Insurance Program: Financial Challenges
Underscore Need for Improved Oversight of Mitigation Programs and Key
Contracts' which was released on July 16, 2008.
This text file was formatted by the U.S. Government Accountability
Office (GAO) to be accessible to users with visual impairments, as part
of a longer term project to improve GAO products' accessibility. Every
attempt has been made to maintain the structural and data integrity of
the original printed product. Accessibility features, such as text
descriptions of tables, consecutively numbered footnotes placed at the
end of the file, and the text of agency comment letters, are provided
but may not exactly duplicate the presentation or format of the printed
version. The portable document format (PDF) file is an exact electronic
replica of the printed version. We welcome your feedback. Please E-mail
your comments regarding the contents or accessibility features of this
document to Webmaster@gao.gov.
This is a work of the U.S. government and is not subject to copyright
protection in the United States. It may be reproduced and distributed
in its entirety without further permission from GAO. Because this work
may contain copyrighted images or other material, permission from the
copyright holder may be necessary if you wish to reproduce this
material separately.
Report to the Ranking Member, Committee on Banking, Housing, and Urban
Affairs, U.S. Senate:
United States Government Accountability Office:
GAO:
June 2008:
National Flood Insurance Program:
Financial Challenges Underscore Need for Improved Oversight of
Mitigation Programs and Key Contracts:
GAO-08-437:
GAO Highlights:
Highlights of GAO-08-437, a report to the Ranking Member, Committee on
Banking, Housing, and Urban Affairs, U.S. Senate.
Why GAO Did This Study:
The Federal Emergency Management Agency (FEMA) and its contractors
administer and implement the National Flood Insurance Program (NFIP).
GAO designated NFIP as a high-risk area in March 2006, and as of
December 2007, FEMA owed more than $17.3 billion to the Treasury for
hurricane-related losses. Concerns have been raised about the financial
condition of NFIP and FEMA’s efforts to mitigate losses and monitor
NFIP contractors. This report (1) describes statistical and financial
trends for NFIP from 1997 through 2006, (2) assesses the extent to
which flood-damaged properties were purchased to mitigate risk, and (3)
evaluates procedures for monitoring NFIP-related contracts.
For this study, GAO analyzed financial and statistical data on the NFIP
and its mitigation programs, reviewed documentation of contract
monitoring activities, and interviewed FEMA officials and contractors.
What GAO Found:
The number of federal flood insurance policies in force nationwide
increased 36 percent from 1997 through 2006, but most homeowners at
risk of flooding still lacked such insurance. While average insurance
amounts (per policy) increased 78 percent from 1997 through
2006—consistent with rising home values—the average premium decreased 3
percent from 1997 through 2006, likely driven in part by the increase
in policies sold in moderate- to low-risk areas. Conversely, loss
amounts fluctuated by year, peaking at more than $17.7 billion in 2005.
Seventy-nine percent of the funds paid out through NFIP from 1997
through 2006 were for hurricane-related claims, but the percentages in
individual years varied widely (correlating with hurricane activity).
Finally, the extent of claim payments attributed to repetitive loss
properties (those with two or more claims in a rolling 10-year period)
increased from 1997 through 2006, from $3.7 billion to nearly $8
billion, with the most significant increases resulting from the 2005
Gulf Coast hurricanes.
Because of data limitations, GAO was not able to determine the actual
number of properties acquired through FEMA mitigation programs, which
are intended to minimize the damage and financial impact of floods.
Information on completed mitigation projects (which encompass multiple
properties) indicates that about one-third of properties approved for
acquisition from 1997 to 2006 were acquired. However, these data are
limited because they do not include a count of properties acquired in
ongoing projects. Projects may take several years to complete, and FEMA
does not report properties acquired until a project is complete.
Further, FEMA collected property acquisition data (for completed
projects) in an ad hoc manner because FEMA’s grants management system
lacks the capability to record acquisition data. As a result, FEMA
cannot readily determine the extent to which flood-damaged and
repetitive loss properties have been acquired through its mitigation
programs.
Lack of monitoring records, inconsistent application of procedures, and
lack of coordination have diminished the effectiveness of FEMA
monitoring of NFIP-related contracts. While federal internal control
standards state that records should be properly maintained, FEMA did
not consistently follow its monitoring procedures for preparing or
maintaining monitoring reports and was unable to provide copies of the
majority of monitoring reports GAO requested. Further, FEMA offices did
not coordinate information and actions relating to contractor
deficiencies and payments. In some cases, key officials were unaware of
decisions on contractor performance. As a result, FEMA cannot
consistently ensure adherence to contract requirements and lacks
information critical for effective oversight of key contractors. Given
the reliance of NFIP upon contractors, it is important that FEMA have
in place adequate controls that are consistently applied to all
contracts.
What GAO Recommends:
GAO recommends that FEMA establish processes to track property
acquisitions in real time and ensure systematic monitoring and reviews
of contractor performance. FEMA agreed with the recommendations on
contractor oversight and has taken steps to address them. FEMA
questioned the value of collecting property acquisition data real-time.
Without such data, FEMA’s ability to assess program effectiveness is
limited.
To view the full product, including the scope and methodology, click on
[hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-08-437]. For more
information, contact Orice M. Williams at (202) 512-8678 or
williamso@gao.gov.
[End of section]
Contents:
Letter:
Results in Brief:
Background:
Policies and Insurance Amounts Increased, Premiums Decreased, and
Losses Fluctuated from 1997 through 2006 with Hurricane Activities:
Funding for Mitigation Programs Has Increased, but FEMA Cannot Readily
Track Real-time Property Acquisitions:
FEMA's Oversight of Contractor Activities Lacked Consistency,
Documentation, and Coordination:
Conclusions:
Recommendations for Executive Action:
Agency Comments and Our Evaluation:
Appendix I: Objectives, Scope, and Methodology:
Appendix II: National Flood Insurance Program Statistics, 1997-2006:
Appendix III: National Flood Insurance Program Statistics by Flood
Zone, 1997-2006:
Appendix IV: National Flood Insurance Program Statistics by Occupancy
Type, 1997-2006:
Appendix V: National Flood Insurance Program Statistics for Hurricane-
Related Losses and Repetitive Loss Properties:
Appendix VI: Flood-Related Damages:
Appendix VII: Definitions of Flood Zones:
Appendix VIII: Comments from the Federal Emergency Management Agency:
Appendix IX: GAO Contact and Staff Acknowledgments:
Tables:
Table 1: NFIP Flood Zone Designations:
Table 2: NFIP-Funded Mitigation Grant Programs:
Table 3: Average Amount of Insurance Coverage per Policy in Force and
Average Premium Paid, 1997-2006:
Table 4: Funding for the FMA Program, 1997-2007:
Table 5: Properties Approved for Acquisition and Acquired through the
FMA Program, 1997-2007 (Acquisition-Only Projects):
Table 6: FMA Program Properties Approved for Acquisition and Acquired,
1997-2007 (Mixed Mitigation Projects):
Table 7: Extent to Which FEMA Produced Required Monthly Monitoring
Reports for the BSA Contractor (October 2005-May 2007):
Table 8: Analysis of FEMA Monthly Monitoring Reports Required for the
BSA Contract (October 2005-May 2007):
Table 9: Number and Percentage Change for Policies in Force, 1997-2006:
Table 10: Maximum Level of Flood Insurance Available, Regular Program,
1997-2006:
Table 11: Average Amount of Insurance Coverage per Policy in Force,
1997-2006:
Table 12: NFIP Average Premium per Policy in Force, 1997-2006:
Table 13: NFIP Losses, 1997-2006:
Table 14: Average Loss per Policy in Force in Dollars, 1997-2006:
Table 15: Average NFIP Claim Payment by Insurance Policy Type, 1997-
2006:
Table 16: Average Historical Loss Year Estimate for NFIP, 1997-2005:
Table 17: NFIP Flood Zone Designations:
Table 18: Number of Policies in Force by Flood Zone, 1997-2006:
Table 19: NFIP Average Amount of Insurance Coverage per Policy in Force
by Flood Zones, 1997-2006:
Table 20: Average Premium Per Policy in Force by Flood Zone, 1997-2006:
Table 21: NFIP Losses Paid Out by Flood Zones, 1997-2006:
Table 22: NFIP Residential (Single-and Two-Four Family) Average Loss
per Claim by Flood Zone, 1997-2006:
Table 23: NFIP Policies in Force--Percentage for Residential (Single-
and Two-Four Family) Properties and Condominiums, 1997-2006:
Table 24: Average Amount of Insurance Coverage by Occupancy Type, 1997-
2006:
Table 25: Average Premium by Occupancy Type, 1997-2006:
Table 26: Losses by Occupancy Type, 1997-2006:
Table 27: NFIP Average Loss per Claim by Occupancy Type, 1997-2006:
Table 28: Percentage of NFIP Losses Attributable to Hurricanes, 1997-
2006:
Table 29: Net Cumulative Total Repetitive Loss (RL) Properties, Net
Cumulative Number of Losses for RL Properties, and Net Cumulative Total
Losses Paid Out for RL Properties, 1997-2006:
Table 30: Percentage of Claims for Which Damage Data Are Available, for
Policies Purchased at the Insurance Limit, 1997-2004:
Table 31: Percentage of Claims for Which Damage Data Are Available, for
Policies Purchased at the Insurance Limit, 2005:
Table 32: Percentage of Claims for Which Damage Data Are Available, for
Policies Purchased at Less Than the Insurance Limit, 1997-2004:
Table 33: Percentage of Claims for Which Damage Data Are Available, for
Policies Purchased at Less Than the Insurance Limit, 2005:
Table 34: Percentage of Claims and Damages in Excess of NFIP Insurance
Available by Policy Type, for Policies Purchased at the Insurance
Limit, 1997-2004:
Table 35: Percentage of Claims and Damages in Excess of NFIP Insurance
Available by Policy Type, for Policies Purchased at the Insurance
Limit, 2005:
Table 36: Percentage of Claims and Damages in Excess of NFIP Insurance
Purchased by Policy Type, for Policies Purchased at Less Than the
Insurance Limit, 1997-2004:
Table 37: Percentage of Claims and Damages in Excess of NFIP Insurance
Purchased by Policy Type, for Policies Purchased at Less Than the
Insurance Limit, 2005:
Figures:
Figure 1: FEMA Offices with Contract Monitoring Responsibilities:
Figure 2: Number of NFIP Policies in Force and Year to Year Percentage
Changes, 1997-2006:
Figure 3: NFIP Average Premium per Policy in Force, by Flood Zone, 1997-
2006:
Figure 4: NFIP Losses, 1997-2006:
Figure 5: Percentage of NFIP Losses Attributable to Hurricanes, 1997-
2006:
Figure 6: Net Cumulative Claims Filed and Losses Paid for Repetitive
Loss Properties, and Number of Repetitive Loss Properties, 1997-2006:
Figure 7: Trends in Percentage Change and Average Amount of Insurance
Coverage per Policy in Force, 1997-2006:
Figure 8: Number of NFIP Claims and Average Loss per Claim, 1997-2006:
Figure 9: Number of Policies in Force by Flood Zone, 1997-2006:
Figure 10: NFIP Average Amount of Insurance Coverage per Policy in
Force in Selected Flood Zones, 1997-2006:
Figure 11: NFIP Losses Paid Out by Flood Zones, 1997-2006:
Figure 12:NFIP Residential (Single-and Two-Four Family) Average Loss
per Claim by Flood Zone for Selected Flood Zones, 1997-2006:
Figure 13: Percentage of Losses Attributable to Residential (Single-and
Two-Four Family) Properties and Condominiums, 1997-2006:
Figure 14:NFIP Average Loss per Claim for Residential (Single-and Two-
Four Family) Properties and Condominiums, 1997-2006:
Abbreviations:
BFE: Base Flood Elevation:
BSA: Bureau and Statistical Agent:
COTR: Contract Officer's Technical Representative:
DHS: Department of Homeland Security:
DSA: Direct Servicing Agent:
eGovernment: Electronic Government:
eGrants: Electronic Grants Management System:
FEMA: Federal Emergency Management Agency:
FIRM: Flood Insurance Rate Map:
FIS: Flood Insurance Study:
FMA: Flood Mitigation Assistance Program:
HMGP: Hazard Mitigation Grant Program:
IFMIS: Integrated Financial Management Information System:
NEMIS: National Emergency Management Information System:
NFIF: National Flood Insurance Fund:
NFIP: National Food Insurance Program:
OMB: Office of Management and Budget:
RFC: Repetitive Flood Claims Program:
RCBAP: Residential Condominium Building Association Policy:
SRL: Severe Repetitive Loss Pilot Program:
TRRP: Transaction Record Reporting and Processing:
WYO: Write Your Own:
[End of section]
United States Government Accountability Office:
Washington, DC 20548:
June 16, 2008:
The Honorable Richard C. Shelby:
Ranking Member:
Committee on Banking, Housing, and Urban Affairs:
United States Senate:
Dear Senator Shelby:
Following the signing of the Bunning-Bereuter-Blumenauer Flood
Insurance Reform Act of 2004, which reauthorized the National Flood
Insurance Program (NFIP) through September 30, 2008, the United States
experienced back-to-back catastrophic hurricane seasons and experienced
nearly $20 billion in flood-insurance losses. Because NFIP premium
rates have been set to cover losses in an average year (that is, based
on "average historical losses"), the program has been unable to set
aside sufficient funds to meet future expected losses.[Footnote 1]
According to the Federal Emergency Management Agency (FEMA) of the
Department of Homeland Security (DHS), which administers NFIP, claims
from Hurricane Katrina alone have totaled in excess of $16.3 billion,
as of October 2007. The Katrina-related claims represent more than
eight times the amount for claims in 2004 and surpass (by more than $1
billion) the aggregate amount of all claims previously paid in the
nearly 40-year history of NFIP.[Footnote 2]
In response to the magnitude and severity of losses from the 2005
hurricanes, Congress increased NFIP's borrowing authority from the
Department of the Treasury (Treasury) to $20.8 billion. As of December
2007, FEMA owed more than $17.3 billion to Treasury, an amount the
program is unlikely to be able to repay while paying future claims with
its current premium income of about $2 billion annually. We designated
NFIP as a high-risk area in March 2006 because the program likely will
not generate sufficient revenues to repay the billions it borrowed from
the Treasury to cover flood claims from the 2005 hurricanes.[Footnote
3] In addition, it is unlikely that NFIP--a key component of the
federal government's efforts to minimize the damage and financial
impact of floods--could cover catastrophic losses in future years. The
insufficient revenues highlight structural weaknesses in how the
program is funded, and Congress is considering a number of legislative
changes to improve NFIP's financial solvency.[Footnote 4]
NFIP was created in 1968 in part to provide some insurance protection
for flood victims because private insurers were (and largely still are)
unwilling to bear the economic risks associated with the potentially
catastrophic impact of flooding. NFIP also provides incentives for
communities to adopt and enforce floodplain management regulations to
reduce future flood damage. Such incentives include mitigation programs
that provide grants to acquire and demolish flood-prone structures that
NFIP insures and perform other activities (such as elevating buildings)
to reduce or eliminate flood losses and insurance claims. Floods are
the most common and destructive natural disasters in the United States.
According to NFIP statistics, 90 percent of all natural disasters in
the United States involve flooding. In previous work, we reported that
between 1980 and 2005, about 97 percent of the U.S. population lived in
a county that had at least one declared flood disaster, and 45 percent
lived in a county that had six or more flood disaster declarations.
[Footnote 5] Under NFIP, federally backed flood insurance is available
to homeowners and other property owners in communities that adopt and
enforce local floodplain ordinances. Finally, FEMA makes extensive use
of contractors to implement NFIP. For example, about 68 FEMA employees,
assisted by about 170 contract employees, manage and oversee NFIP.
As agreed with your staff, this report:
* describes trends for NFIP policies, insurance amounts, premiums, and
losses from 1997 through 2006, and the extent to which NFIP losses were
attributable to hurricanes and repetitive loss properties;
* assesses how the amounts available for the purchase of flood-damaged
and repetitive loss properties changed over time, and the extent to
which FEMA purchased flood-damaged and repetitive loss properties; and:
* evaluates the extent to which FEMA followed its procedures for
monitoring selected NFIP-related contracts.
To address these objectives, we analyzed data on NFIP policies,
insurance amounts, premiums, and losses for calendar years 1997 through
2006 and assessed the reliability of these data. In addition, we
analyzed available data on funds available for three NFIP-funded
mitigation programs and properties purchased in connection with two of
these programs. We reviewed monthly contractor monitoring reports and
documentation of contract-specific monitoring policies and procedures
at FEMA. We also assessed the extent to which these procedures were
followed for two of the largest NFIP-related contracts--the Bureau and
Statistical Agent (BSA) and Direct Servicing Agent (DSA) contracts--
upon which FEMA relies to collect all financial and statistical data
related to NFIP and directly sell and service flood insurance policies.
Finally, we interviewed FEMA officials in Washington, D.C., and
Virginia and NFIP contractors based in Kansas and Maryland. Appendix I
provides more information on our scope and methodology. We conducted
this performance audit from March 2007 to June 2008 in accordance with
generally accepted government auditing standards. Those standards
require that we plan and perform the audit to obtain sufficient,
appropriate evidence to provide a reasonable basis for our findings and
conclusions based on our audit objectives. We believe that the evidence
obtained provides a reasonable basis for our findings and conclusions
based on our audit objectives.
Results in Brief:
From 1997 through 2006, NFIP showed steady increases in the number of
policies in force, the average amount of insurance per policy, and the
net cumulative amount paid to settle claims for repetitive loss
properties; while average premiums decreased, total losses fluctuated,
and the majority of losses were hurricane-related. The number of
policies in force increased 36 percent, from nearly 4 million in 1997
to more than 5.4 million in 2006. However, a 2006 study estimated that
about half of single-family homes in special flood hazard areas
nationwide did not have flood insurance policies. The average amount of
insurance per policy increased 78 percent from 1997 through 2006, from
about $158,000 to almost $214,000.[Footnote 6] The average annual
premium per flood insurance policy decreased slightly from 1997 through
2006, from $489 to $475 annually in part because most of the policies
being sold were in moderate-and low-risk areas that tend to have lower
rates. However, premiums increased in the high-risk coastal flood zone
from about $1,039 in 1997 to more than $1,400 in 2006. Total loss
amounts fluctuated from year to year, ranging from a low of $302
million in 2000 to more than $17.7 billion in 2005. Further, 79 percent
of the funds paid out through NFIP from 1997 through 2006 were for
hurricane-related claims, but the percentages in individual years
varied widely (correlating with hurricane activity). More specifically,
while less than 10 percent of the losses paid out in 5 of the 10 years
were linked to hurricane-related damage, 97 percent of the losses paid
out in 2005, the highest loss year in the history of the program, were
for hurricane-related losses. The net cumulative amount FEMA had paid
to settle claims filed for repetitive loss properties (properties with
two or more claims paid by NFIP in a rolling 10-year period) totaled
$7.9 billion at the end of 2006, more than doubling from the net
cumulative total of $3.7 billion in 1997. Clearly, repetitive loss
properties continue to be a drain on NFIP. FEMA's annual estimates of
average historical losses also increased, from $622 million in calendar
year 1997 to $2.3 billion in 2006. Finally, analysis by occupancy type
shows that more than 90 percent of policyholders owned residential
properties (single-and two-four family homes) or condominiums, and the
majority of these properties were in flood zones that FEMA designated
as high risk.
The amounts available for mitigation activities such as the purchase
and demolition of flood-damaged properties increased for one of NFIP's
mitigation programs and became available for two others in 2006.
Properties have been purchased through two of the three NFIP-funded
mitigation programs--Flood Mitigation Assistance (FMA) and Repetitive
Flood Claims (RFC). However, the total number of properties purchased
is unknown at the headquarters level due to a lack of reporting
requirements and information system limitations that hamper assessments
of the extent of property acquisitions in the two programs. According
to data provided by FEMA, funding for the FMA program increased in the
period we reviewed, from $12.5 million in fiscal year 1997 to nearly
$31 million in fiscal year 2006, while funds for the other two
programs--RFC and the Severe Repetitive Loss (SRL) pilot program--only
have been available since 2006. In addition, the SRL pilot program
guidance was not published and applications for grant funds were not
accepted until January 2008.[Footnote 7] According to available data on
completed mitigation projects, 35 percent of properties approved for
acquisition through FMA, and 19 percent of properties approved through
RFC, have been acquired as of October 2007. However, the total number
of properties acquired through NFIP-funded mitigation programs is
unknown because FEMA lacks real-time data on property acquisitions for
ongoing mitigation projects. For example, of the properties reported as
being acquired through the FMA program as of October 2007, FEMA records
show that none have been acquired since 2003. However, this number is
likely understated because FEMA does not explicitly require the timely
recording of property acquisitions, and FEMA regional staff may not
record property acquisitions until a project closes, which could take
several years. Internal control standards for the federal government
direct that transactions be promptly recorded to maintain their
relevance and value to management in controlling operations and making
decisions.[Footnote 8] In addition, FEMA's grants management system
does not allow for the electronic capture of data on property
acquisitions. As a result of these limitations, FEMA management cannot
readily determine the effectiveness of the mitigation programs and the
number of acquisitions, a key means of reducing or eliminating future
insurance claims on NFIP-insured properties and determining the
effectiveness of these programs.
FEMA lacked monitoring records, inconsistently followed its procedures
for monitoring contractors, and did not coordinate contract monitoring
responsibilities for the two major contracts we reviewed. At FEMA, a
Contracting Officer's Technical Representative (COTR) and staff
(referred to as "monitors") are responsible, respectively, for ensuring
compliance with contract terms and regularly monitoring and reporting
on the extent to which an NFIP contractor has met standards in contract-
specified performance areas. FEMA lacked records for the majority of
the monitoring reports we requested and did not consistently follow the
monitoring procedures for preparing, reviewing, and maintaining
monitoring reports. Internal control standards for the federal
government state that records should be properly managed and
maintained. Additionally, key FEMA offices responsible for addressing
contractor deficiencies did not share information and coordinate
actions about contractor performance problems and payments. As a
result, FEMA cannot ensure adherence to contract requirements and lacks
information critical for effective oversight of contractors performing
key NFIP data collection, reporting, and insurance functions.
We are making five recommendations to the Secretary of Homeland
Security to more accurately track the extent to which flood-prone
properties are acquired in NFIP mitigation programs and help ensure
adherence to contract monitoring requirements. They include
establishing written guidance about consistent and timely recording of
property acquisition data and establishing a way to track real-time
property acquisitions. The remaining recommendations address contract
management and oversight and include establishing a process to
systematically review monitoring reports and develop guidance to
improve coordination among key FEMA offices in addressing contractor
deficiencies.
We provided a draft of this report to the Secretary of Homeland
Security. The Assistant Administrator, Mitigation Directorate, FEMA,
provided written comments that are reprinted in appendix VIII. In these
comments, FEMA generally agreed with the three recommendations on
contract monitoring and described various steps they are taking to
implement them. While FEMA did not explicitly address our two
recommendations related to tracking property acquisitions, it noted the
challenges it faces in tracking property acquisitions but questioned
the value of tracking such acquisitions at the headquarters level. As
noted in our report, real-time data would improve FEMA headquarters'
ability to produce, analyze, and report timely information on ongoing
operations and thus improve its ability to assess the effectiveness of
its mitigation programs. Specifically, FEMA would be able to more
accurately assess the rate at which properties are acquired and thus
pinpoint within a shorter timeframe the extent to which mitigation
programs were reducing the number of flood-damaged and repetitive loss
properties. Finally, FEMA's written response also included technical
comments that we incorporated in this report as appropriate.
Background:
NFIP was established in the National Flood Insurance Act of 1968 to
provide policyholders with some insurance coverage for flood damage, as
an alternative to disaster assistance, and to try to reduce the
escalating costs of repairing flood damage.[Footnote 9] NFIP has mapped
flood risks, assigning a flood zone designation based on the risk level
for flooding (see table 1 for examples), which is a factor in
determining premium rates.
Table 1: NFIP Flood Zone Designations:
Designations: Flood zones B,C, X;
Risk level: Moderate-to low-risk.
Designations: Flood zones A, AE;
Risk level: High-risk.
Designations: Flood zones V, VE;
Risk level: High-risk coastal.
Designations: Flood zone D;
Risk level: Undetermined risk.
Source: GAO analysis of FEMA data.
Notes: See appendix VII for a description of all FEMA flood zones. The
FEMA contractor responsible for collection and analysis of NFIP data
uses two additional designations when collecting and reporting data by
flood zone. Zone "O" (Other) is not a flood zone designation; rather,
it is used to indicate missing or erroneous data for policies. Policies
under FEMA's Emergency Program, which is the program through which
communities enter NFIP, do not have designated flood zones. Instead,
the FEMA contractor captures data on the policies by using their
Emergency Program status.
[End of table]
To participate in NFIP, communities agree to enforce regulations for
land use and new construction in high-risk flood zones. As of May 2007,
more than 20,300 communities across the United States and its
territories participated in NFIP by adopting and agreeing to enforce
state and community floodplain management regulations to reduce future
flood damage.[Footnote 10] Although community participation in the
program is voluntary, homeowners with mortgages from federally
regulated lenders on properties in special high-risk, flood hazard
areas are required to purchase flood insurance on their dwellings for
at least the outstanding mortgage amount, up to the maximum policy
limit of $250,000.[Footnote 11] Optional, lower-cost coverage also is
available under NFIP to protect homes in areas of moderate-to low-risk.
In addition, owners of properties built before NFIP officially mapped a
community (known as pre-FIRM properties, referring to the Flood
Insurance Rate Map) pay premiums that do not reflect their true risk
because Congress authorized subsidized insurance rates to be made
available for policies covering certain structures to encourage
communities to join the program. To insure other personal property
items against flood damage, homeowners may purchase separate NFIP
personal property (or "contents") coverage, up to the maximum coverage
limit of $100,000.[Footnote 12]
In addition to mapping and categorizing flood risk zones, FEMA also has
developed two estimates of potential losses to help set rates and
develop information relating to losses. FEMA introduced the concept of
the average historical loss year--the purpose of which is to estimate
the amount of premium that would be sufficient to pay for losses
resulting from the type of loss years that previously occurred.
According to FEMA, this estimate is used as one indicator in setting
subsidized premium rates. FEMA also prepares a catastrophic loss year
estimate, usually every other year, to provide Congress with an
informal guide on the losses that could occur in the event of a storm
that would have 0.1 percent chance of occurring in a given year (a 1 in
a 1,000 year event).
For more than a decade, FEMA has pursued a variety of strategies to
reduce the number of repetitive loss properties, which are insurable
buildings for which NFIP paid two or more claims of more than $1,000
within any rolling 10-year period. For example, NFIP funds three
mitigation programs, the FMA program, the RFC program, and the SRL
pilot program.[Footnote 13] The purpose of the FMA program is to fund
activities that reduce or eliminate the long-term risk of flood damage
to buildings and other structures insured by NFIP, such as property
acquisition.[Footnote 14] The FMA program requires applicants to
provide a portion of the funds for the proposed mitigation activities.
If the applicant cannot secure funding for the proposed activities,
FEMA may contribute 100 percent of the funding through the RFC program,
which is specifically designed to mitigate repetitive loss properties
(structures insured by NFIP that have had one or more claim payments
for flood damages).[Footnote 15] Finally, the SRL program is designed
for repetitive loss properties that have at least four claim payments
of more than $5,000 each.[Footnote 16] See table 2 for more information
on these mitigation programs.
Table 2: NFIP-Funded Mitigation Grant Programs:
Program element: Purpose;
FMA: To implement cost-effective measures that reduce or eliminate the
long-term risk of flood damage to buildings, manufactured homes, and
other structures insured under NFIP;
RFC: To reduce or eliminate the long-term risk of flood damage to
structures insured under NFIP that have had one or more claim payments
for flood damage;
SRL: To reduce or eliminate the long-term risk of flood damage to
severe repetitive loss residential properties and the associated drain
on the National Flood Insurance Fund (NFIF) from such properties.
Program element: Applicant eligibility;
FMA: State emergency management agencies or a similar state office[A];
RFC: Same as FMA, but for those states or communities that cannot meet
the requirements of the FMA program for either cost share or capacity
to manage the activities;
SRL: State emergency management agencies or a similar state office.[A]
Program element: Eligible project grants;
FMA: Acquisition, structure demolition, or structure relocation with
the property deed restricted for open space uses in perpetuity;
elevation of structures; dry floodproofing of nonresidential
structures; and minor structural flood control activities;
RFC: Acquisition, structure demolition, or structure relocation with
the property deed restricted for open space uses in perpetuity;
SRL: Acquisition, structure demolition, or structure relocation with
the property deed restricted for open space uses in perpetuity;
elevation of structures; dry floodproofing of historic structures;
minor physical localized flood control projects; and mitigation
reconstruction (demolition and rebuilding of structures).
Program element: Funding;
FMA: 2-year funds[B];
RFC: No-year funds;
SRL: No-year funds.
Source: FEMA.
Note: In fiscal year 2008, appropriations for RFC changed from "one-
year funds"--congressionally appropriated funds that must be expended
within the fiscal year--to "no-year" funds, congressionally
appropriated funds with no time limit for expenditure.
[A] The office that has primary emergency management or floodplain
management responsibility.
[B] Congressionally appropriated funds that must be expended within two
fiscal years.
[End of table]
NFIP staff are located at FEMA headquarters in Washington, D.C., and 10
regional offices. Their functions include tasks for administering
mitigation grant programs such as processing applications for grant
funds and distributing the funds to eligible applicants. While FEMA
establishes policies and procedures for NFIP, and has some staff
dedicated to the management of the program, FEMA relies on contractors
to administer key aspects of the program. Contractors collect NFIP
data, market the program, and sell specific flood-insurance policies.
Private insurance companies that participate in FEMA's Write Your Own
(WYO) program largely are responsible for insurance sales and claims
adjustment.[Footnote 17]
At the time of our work, there were four major ongoing contracts in
place for NFIP, two of which we reviewed in this report. For example,
FEMA awards a contract for a BSA, which is responsible for conducting
financial and statistical reporting based upon data submissions from
the 92 WYO companies, developing forms and information related to NFIP,
and providing various data analyses.[Footnote 18] The BSA contract that
was effective from September 2005 through December 2007 was valued at
$38 million. FEMA also awards a contract for a DSA, which is
responsible for selling and servicing standard and group insurance for
policies that are not sold through WYOs. For that agent, the value of
the contract in effect from September 2003 to September 2008 is $26.5
million.
Two offices in FEMA are responsible for contracts and contract
oversight (see fig. 1). A Contracting Officer in the Acquisition
Operations Branch awards contracts related to NFIP and has
responsibility for determining the adequacy of a contractor's
performance.[Footnote 19] The Program Management Office also has
responsibilities for overseeing each contractor's performance. A COTR
is assigned to each contract. The COTR is authorized to act on behalf
of the Contracting Officer to monitor the contract. However, the COTR
is not authorized to make any contractual commitments or contractual
changes on behalf of FEMA; any changes that the contractor deems may
affect the contract, price, terms, or conditions are referred to the
Contracting Officer in the Acquisition Operations Branch for action.
Each COTR is assisted by monitors, who review and report on technical
aspects of the contractor's performance such as compliance with
specific contract terms.
Figure 1: FEMA Offices with Contract Monitoring Responsibilities:
[See PDF for image]
This figure is an illustration of FEMA offices with contract monitoring
responsibilities, as follows:
Acquisition Operations Branch:
* Mitigation Section;
- Contracting Officer (enters into, administers, and terminates
contracts).
Mitigation Directorate:
* Risk Insurance Division:
- Operations Management Branch:
- COTR (assure proper surveillance of the contractor’s performance);
- Monitor (responsible for the technical administration of specific
functional areas under the contract).
Source: GAO.
Note: During the period of our review, the Operations Management Branch
was referred to as the Program Management Office.
[End of figure]
Policies and Insurance Amounts Increased, Premiums Decreased, and
Losses Fluctuated from 1997 through 2006 with Hurricane Activities:
Overall, data relating to the number of NFIP policies and amount of
insurance coverage indicated upward trends for the period we reviewed
and most losses were paid for hurricane-related claims, primarily due
to losses from the 2005 hurricane season. The number of policies in
force increased 36 percent and the average amount of coverage increased
78 percent from 1997 through 2006, consistent with average rising home
values. The average premium per flood-insurance policy decreased 3
percent, but when analyzed by flood risk zone, average premiums
increased in the high-risk coastal flood zone, from about $1,039 in
1997 to more than $1,400 in 2006. Over this period, total losses
fluctuated from a low of $302 million in 2000 to a high of $2 billion
in 2004 before reaching more than $17.7 billion in 2005. In 5 of the 10
years reviewed, less than 10 percent of losses paid out were linked to
hurricane-related damage, and over the 10-year period cumulative losses
for repetitive loss properties increased from $3.7 billion to $7.9
billion. FEMA's estimates for average historical loss and catastrophic
loss increased over this 10-year period. Finally, analysis by occupancy
type shows more than 90 percent of policyholders owned residential
properties (single and two-four family homes) or condominiums, and the
majority of these properties were found in zones that FEMA designated
as high risk. For additional statistical and trend data, see appendixes
II-VI.
Policies in Force and Average Insurance Amounts Increased, and Average
Premiums Decreased:
The number of federal flood insurance policies in force increased from
3.96 million in 1997 to 5.40 million at the end of 2006, an increase of
36 percent.[Footnote 20] The number of policies increased from 1 to 6
percent each year from 1997 through 2005, but increased 12 percent the
year after the Gulf Coast hurricanes (see fig. 2). While the number of
policies in force has increased, a 2006 study commissioned by FEMA
estimated that about half of the single-family homes in special flood
hazard areas nationwide do not have flood insurance policies.[Footnote
21] Moreover, the study estimated that market penetration outside of
special flood hazard areas is about 1 percent.
Figure 2: Number of NFIP Policies in Force and Year to Year Percentage
Changes, 1997-2006:
[See PDF for image]
This figure is a combination line and vertical bar graph depicting the
following data:
Calendar year: 1997;
Policies in force: 3,962,080;
Percentage change (from year to year): baseline year.
Calendar year: 1998;
Policies in force: 4,114,32o;
Percentage change (from year to year): +4%.
Calendar year: 1999;
Policies in force: 4,206,910;
Percentage change (from year to year): +2%.
Calendar year: 2000;
Policies in force: 4,255,430;
Percentage change (from year to year): +1%.
Calendar year: 2001;
Policies in force: 4,360,680;
Percentage change (from year to year): +2%.
Calendar year: 2002;
Policies in force: 4,406,660;
Percentage change (from year to year): +1%.
Calendar year: 2003;
Policies in force: 4,447,770;
Percentage change (from year to year): +1%.
Calendar year: 2004;
Policies in force: 4,558,700;
Percentage change (from year to year): +2%.
Calendar year: 2005;
Policies in force: 4,827,180;
Percentage change (from year to year): +6%.
Calendar year: 2006;
Policies in force: 5,404,950;
Percentage change (from year to year): +12%.
Source: GAO analysis of NFIP data.
Note: See appendix II for more detailed information.
[End of figure]
The average amount of insurance coverage per policy in force increased
from about $158,000 in 1997 to almost $214,000 at the end of 2006, an
increase of 78 percent (see table 3).[Footnote 22] While we did not
attempt to determine the reason that the average amount of insurance
purchased increased, we note that this increase coincides with the
increase in average home values over this period.[Footnote 23]
Table 3: Average Amount of Insurance Coverage per Policy in Force and
Average Premium Paid, 1997-2006:
Calendar year: 1997;
Average insurance per policy in force: $158,125;
Average premium paid: $489.
Calendar year: 1998;
Average insurance per policy in force: $163,688;
Average premium paid: $503.
Calendar year: 1999;
Average insurance per policy in force: $167,162;
Average premium paid: $493.
Calendar year: 2000;
Average insurance per policy in force: $168,788;
Average premium paid: $468.
Calendar year: 2001;
Average insurance per policy in force: $176,797;
Average premium paid: $451.
Calendar year: 2002;
Average insurance per policy in force: $178,467;
Average premium paid: $444.
Calendar year: 2003;
Average insurance per policy in force: $187,709;
Average premium paid: $453.
Calendar year: 2004;
Average insurance per policy in force: $198,136;
Average premium paid: $465.
Calendar year: 2005;
Average insurance per policy in force: $206,501;
Average premium paid: $469.
Calendar year: 2006;
Average insurance per policy in force: $213,944;
Average premium paid: $475.
Source: GAO analysis of NFIP data.
Note: Dollars are in 2006 constant dollars.
[End of table]
The average annual premium per policy in force decreased from $489 in
1997 to $475 at the end of 2006, a decrease of almost 3 percent (see
table 3). According to FEMA, the average premium has been influenced by
changes in two types of flood zones. First, the average premium has
been strongly influenced by changes in the moderate-to low-risk flood
zones, where policies as a percentage of the total policies in force
have increased greatly primarily due to increased sales of Preferred
Risk Policies, FEMA's lowest-cost flood insurance policy.[Footnote 24]
Second, average annual premium per policy has been influenced by the
steep increase in the average premium of high-risk coastal zone
policies, but because they make up a smaller percent of policies in
force, this increase did not drive up the average policy premium. As
figure 3 illustrates, the average premium increased in the high-risk
coastal zone the most (41 percent), from about $1,039 in 1997 to more
than $1,400 at the end of 2006.
Figure 3: NFIP Average Premium per Policy in Force, by Flood Zone, 1997-
2006:
[See PDF for image]
This figure is a multiple line graph depicting the following data:
NFIP Average Premium per Policy in Force, by Flood Zone, 1997-2006:
Calendar year: 1997;
High-risk (Zone A): $520.7;
High-risk coastal (Zone V): $1038.6;
Moderate- to low-risk (Zones B, C, and X): $377.05.
Calendar year: 1998;
High-risk (Zone A): $537.82;
High-risk coastal (Zone V): $1041.1;
Moderate- to low-risk (Zones B, C, and X): $384.43.
Calendar year: 1999;
High-risk (Zone A): $525.27;
High-risk coastal (Zone V): $1053.33;
Moderate- to low-risk (Zones B, C, and X): $379.67.
Calendar year: 2000;
High-risk (Zone A): $497.2;
High-risk coastal (Zone V): $1049.2;
Moderate- to low-risk (Zones B, C, and X): $363.24.
Calendar year: 2001;
High-risk (Zone A): $481.58;
High-risk coastal (Zone V): $1059.18;
Moderate- to low-risk (Zones B, C, and X): $349.61.
Calendar year: 2002;
High-risk (Zone A): $471.68;
High-risk coastal (Zone V): $1117.63;
Moderate- to low-risk (Zones B, C, and X): $343.24.
Calendar year: 2003;
High-risk (Zone A): $478.47;
High-risk coastal (Zone V): $1231.85;
Moderate- to low-risk (Zones B, C, and X): $349.38.
Calendar year: 2004;
High-risk (Zone A): $491.96;
High-risk coastal (Zone V): $1343.04;
Moderate- to low-risk (Zones B, C, and X): $343.81.
Calendar year: 2005;
High-risk (Zone A): $502.28;
High-risk coastal (Zone V): $1398.43;
Moderate- to low-risk (Zones B, C, and X): $339.52.
Calendar year: 2006;
High-risk (Zone A): $520.57;
High-risk coastal (Zone V): $1463.34;
Moderate- to low-risk (Zones B, C, and X): $335.97.
Source: GAO analysis of NFIP data.
Notes: We excluded flood zone D and the Emergency Program because
policies in force in these zones represented less than 1 percent of the
policies in force for each year reviewed. Dollars are in 2006 constant
dollars. See appendix III for supporting data.
[End of figure]
Total Losses Fluctuated in the Period We Reviewed, and the Majority of
Losses Were Attributable to Hurricanes:
With the exception of 2005, the amount that FEMA paid out in losses
over the 10-year period fluctuated between a low of $302 million in
2000 and a high of $2 billion in 2004 (see fig. 4).[Footnote 25] Losses
refer to the amount that FEMA paid to settle flood insurance claims and
are recorded for the year in which the loss occurred.[Footnote 26] For
example, if a flood damaged an insured property in 2005, but the claim
was not paid until 2006, the amount paid out would be recorded as a
2005 loss. For 2005, as a result of the Gulf Coast hurricanes, FEMA
paid out more than $17.7 billion in losses.[Footnote 27]
Figure 4: NFIP Losses, 1997-2006:
[See PDF for image]
This figure is a vertical bar graph depicting the following data:
NFIP Losses, 1997-2006:
Calendar year (number of hurricanes): 1997 (2);
Losses: $0.684 billion;
Calendar year (number of hurricanes): 1998 (9);
Losses: $1.129 billion;
Calendar year (number of hurricanes): 1999 (2);
Losses: $0.935 billion;
Calendar year (number of hurricanes): 2000 (1);
Losses: $0.302 billion;
Calendar year (number of hurricanes): 2001 (2);
Losses: $1.472 billion;
Calendar year (number of hurricanes): 2002 (4);
Losses: $0.48 billion;
Calendar year (number of hurricanes): 2003 (1);
Losses: $0.82 billion;
Calendar year (number of hurricanes): 2004 (4);
Losses: $2.273 billion;
Calendar year (number of hurricanes): 2005 (5);
Losses: $17.71 billion;
Calendar year (number of hurricanes): 2006 (2);
Losses: $0.612 billion.
Source: GAO analysis of NFIP data.
Notes: "Other residential" and "nonresidential" are excluded because,
for each year of our review, over 90 percent of all policies were for
the residential (single-and two-four family) and the condominium
occupancy type. Dollars are in 2006 constant dollars. We excluded loss-
related expenses because they were not available for all of the
categories of losses in our analysis, such as hurricane-related losses.
Loss-related expenses, which accounted for from 6-8 percent of the
total losses paid out each year from 1997 through 2006, refer to the
administrative costs associated with paying losses. See appendix II for
supporting data.
[End of figure]
From 1997 through 2006, 79 percent of the funds paid out in losses were
for hurricane-related claims, but the percentages in individual years
varied widely (correlating to hurricane activity). As shown in figure
5, in five of the years reviewed, less than 10 percent of losses were
attributable to hurricanes, and in four other years, more than 60
percent of losses were attributable to hurricanes.[Footnote 28] In
2005, the Gulf Coast hurricanes accounted for 97 percent of all losses
paid. Moreover, the 2005 hurricanes accounted for 65 percent of the
losses paid during the 10-year period we reviewed. As we have
previously reported, it is highly unlikely that NFIP, as currently
funded, could generate revenues to repay the amount borrowed from
Treasury to pay for claims from the 2005 hurricane season.[Footnote 29]
Figure 5: Percentage of NFIP Losses Attributable to Hurricanes, 1997-
2006:
[See PDF for image]
This figure is a vertical bar graph depicting the following data:
Percentage of NFIP Losses Attributable to Hurricanes, 1997-2006:
Calendar year: 1997;
Percentage of losses: 2.4%.
Calendar year: 1998;
Percentage of losses: 28%.
Calendar year: 1999;
Percentage of losses: 81%.
Calendar year: 2000;
Percentage of losses: 1.1%.
Calendar year: 2001;
Percentage of losses: 0.03%.
Calendar year: 2002;
Percentage of losses: 8.4%.
Calendar year: 2003;
Percentage of losses: 65%.
Calendar year: 2004;
Percentage of losses: 85%.
Calendar year: 2005;
Percentage of losses: 97%.
Calendar year: 2006;
Percentage of losses: 6.4%.
Source: GAO analysis of NFIP data.
Notes: In 2001, approximately $463,000, or less than 1 percent of total
NFIP losses, was attributable to hurricanes. See appendix V for
supporting data.
[End of figure]
From 1997 through 2006, the net cumulative amount that FEMA paid to
settle claims filed for repetitive loss properties increased from $3.7
billion to $7.9 billion.[Footnote 30] Congress has noted that
repetitive loss properties constitute a significant drain upon NFIP.
While FEMA has tried to decrease the portfolio of repetitive loss
properties, during this same period, the net cumulative number of
repetitive loss properties increased approximately 64 percent, from
76,000 to 125,000 (see fig. 6). Clearly, repetitive loss properties
continue to be a problem for NFIP. While these repetitive loss
properties account for about 1 percent of NFIP-insured properties, they
account for between 25 and 30 percent of all flood claims. We could not
determine the extent to which losses paid in a given year were
attributable to repetitive loss properties because FEMA collects net
cumulative, not annual, data on such losses. However, as shown in
figure 6, the cumulative totals have continued to increase, in
particular in the past 3 years, as the number of repetitive loss
properties has increased.
Figure 6: Net Cumulative Claims Filed and Losses Paid for Repetitive
Loss Properties, and Number of Repetitive Loss Properties, 1997-2006:
[See PDF for image]
This figure is a combined vertical bar and line graph depicting the
following data:
Calendar year: 1997;
Number of repetitive loss claims: 207,550;
Number of repetitive loss properties: 76,108;
Total repetitive loss dollars: $3.69 billion.
Calendar year: 1998;
Number of repetitive loss claims: 229,895;
Number of repetitive loss properties: 83,374;
Total repetitive loss dollars: $4.1 billion.
Calendar year: 1999;
Number of repetitive loss claims: 235,649;
Number of repetitive loss properties: 85,234;
Total repetitive loss dollars: $4.17 billion.
Calendar year: 2000;
Number of repetitive loss claims: 254,313;
Number of repetitive loss properties: 91,731;
Total repetitive loss dollars: $4.59 billion.
Calendar year: 2001;
Number of repetitive loss claims: 264,024;
Number of repetitive loss properties: 95,177;
Total repetitive loss dollars: $4.83 billion.
Calendar year: 2002;
Number of repetitive loss claims: 271,945;
Number of repetitive loss properties: 97,881;
Total repetitive loss dollars: $4.9 billion.
Calendar year: 2003;
Number of repetitive loss claims: 281,426;
Number of repetitive loss properties: 101,173;
Total repetitive loss dollars: $5.01 billion.
Calendar year: 2004;
Number of repetitive loss claims: 297,377;
Number of repetitive loss properties: 107,041;
Total repetitive loss dollars: $5.34 billion.
Calendar year: 2005;
Number of repetitive loss claims: 333,756;
Number of repetitive loss properties: 119,292;
Total repetitive loss dollars: $7.23 billion.
Calendar year: 2006;
Number of repetitive loss claims: 354,010;
Number of repetitive loss properties: 125,239;
Total repetitive loss dollars: $7.94 billion.
Source: GA analysis of NFIP data.
Notes: Dollars are in 2006 constant dollars. All claims and loss data
are net cumulative. See appendix V for supporting data.
[End of figure]
FEMA's Average Historical Loss and Catastrophic Loss Estimates
Increased from 1997 through 2005:
FEMA's average historical loss estimate increased from $622 million in
1997 to $1 billion in 2005, or about 62 percent. As previously noted,
FEMA uses this estimate to calculate the premium that would be
sufficient to pay for the average level of losses that occurred in past
years and help set the rate level for subsidized flood insurance
policies. The estimate is also part of the calculation that FEMA uses
to determine the minimum subsidy price that should be charged for pre-
FIRM flood insurance policies. When this calculation is used, it
ensures that sufficient premiums are collected, in aggregate, to cover
at least the average annual loss as determined by historical data (see
app. II).[Footnote 31] This estimate increased from 1997 to 2005
because it is annually updated with the previous years' flooding
activity. To date, FEMA has not determined how it will use the 2005
catastrophic losses from Hurricanes Katrina, Rita, and Wilma in its
2006 or subsequent average historical loss year calculations.
FEMA's estimates (in ranges) for probable maximum loss in the event of
a catastrophic loss year gradually increased from $6-9 billion in 1998
(the first year the estimate was made) to $8-11 billion in 2000, $9-12
billion in 2002, and to $11-15 billion in 2005.[Footnote 32] According
to FEMA, the estimate increased over this period due to inflation and
the increased policyholder base. Estimates of a catastrophic loss year
are designed to provide Congress with an informal guide on the loss
potential of a catastrophic event--an event that has a 0.1 percent
chance of occurring in any given year. According to FEMA, these
estimates generally are done every other year due to the slow growth in
policies in force, limited actuarial staff, and the imprecision of such
estimates.
More Than 90 Percent of Policies Were for Residences (Single and Two-
Four Family) and Condominiums, with Coverage and Premiums Higher for
Residences Than for Condominiums:
Analysis of policies in force, average insurance per policy, average
premium per policy, and losses by two occupancy types--residential and
condominium--from 1997 through 2006 indicates the following:[Footnote
33]
* More than 90 percent of the insured properties annually were
residential (single-and two-four family) or condominium properties from
1997 through 2006. Specifically, more than 70 percent of the insured
properties were residential properties, and more than 20 percent were
condominiums.[Footnote 34]
* For residential properties, the average amount of insurance purchased
per flood insurance policy ranged from about $170,000 in 1997 to about
$235,000 in 2006. For condominiums, the average amount of flood
insurance per policy increased from about $98,000 in 1997 to $121,000
in 2006. As mentioned previously, this is consistent with the increase
in house values during this period.
* For residential properties, the average premium decreased from about
$522 in 1997 to about $494 in 2006. For condominiums, the average
premium decreased from about $216 in 1997 to $201 in 2006. Premiums may
have decreased because the number of policies in the moderate-and low-
risk zones, which have more affordable premiums, have increased at a
greater rate than in other zones--even though the average amount of
insurance was increasing.
* Residential and condominium losses accounted for 82.5 percent and 3
percent, respectively, of all flood insurance losses that occurred from
1997 through 2006. Losses paid out for residential properties ranged
from a low of $185 million in 2000 to a high of $15.5 billion in 2005,
a result of the Gulf Coast hurricanes. Losses paid out for condominiums
ranged from $6.6 million in 2002 to $395 million in 2004, which were
largely driven by the 2004 Florida hurricanes. The average loss per
claim (for building-only, contents-only, and building and contents
policies) was higher for condominiums than for residential properties
because the data for condominium includes losses for entire condominium
structures, not just individual units. For example, if an insured
condominium building flooded, the (building) claim amount would be for
the whole building and not for the average amount per unit.
See appendix IV for additional detail about policies in force by
dwelling type, amount of insurance purchased, average premium, and
losses by occupancy type.
Funding for Mitigation Programs Has Increased, but FEMA Cannot Readily
Track Real-time Property Acquisitions:
Funding for NFIP mitigation programs that can be used to acquire
properties that are at risk of repetitive flooding has increased from
1997 through 2006, but FEMA's ability to track the effectiveness of
these programs is limited because it does not track property
acquisitions real time. As we mentioned earlier, property acquisition
is one of the tools that FEMA uses to reduce losses from flood-damaged
and repetitive loss properties. FEMA acquires these properties out of
funds appropriated for mitigation programs.[Footnote 35] Three of these
programs are funded by the NFIP: the FMA program, the RFC program, and
the SRL pilot program. Since 1997, annual funding for FMA more than
doubled, while funding for the RFC program and the SRL pilot program
remained the same from 2006, the first year of funding, to 2007.
[Footnote 36] While overall funding for mitigation has increased, the
total number of properties that have been acquired through the
mitigation programs that have expended funds--the FMA and RFC programs-
-is unknown. As a result, FEMA's ability to evaluate the effectiveness
of the programs, such as the extent to which repetitive loss properties
have been acquired, is limited. For the FMA and RFC programs, FEMA does
not have real-time property acquisition data because the agency does
not require its regional staff to report the status of individual
property-level acquisitions to headquarters before a project closes--a
process that can take several years--and its current grants management
system does not have the capability to capture and track such data. As
a result, FEMA management cannot readily access information on the
current status of property acquisitions, limiting their ability to
evaluate the effectiveness of these programs on a real-time basis.
Funds for One Mitigation Program Increased, While Funds for Two Other
Mitigation Programs Only Have Been Available Since 2006:
The purpose of the FMA program is to fund activities that reduce or
eliminate the long-term risk of flood damage to buildings and other
structures insured by NFIP. FEMA data shows that, for the FMA program,
funding increased from $12.5 million in fiscal year 1997 to $31 million
in fiscal year 2007 (see table 4). During this period, nearly $229
million was appropriated to reduce or eliminate the long-term risk of
flood damage to buildings, manufactured homes, and other structures
insured under NFIP.
Table 4: Funding for the FMA Program, 1997-2007:
Fiscal year: 1997;
Amount appropriated: $12,500,000;
Obligational authority brought forward[A]: $7,500,000.
Fiscal year: 1998;
Amount appropriated: $17,500,000;
Obligational authority brought forward[A]: $6,917,400.
Fiscal year: 1999;
Amount appropriated: $20,000,000;
Obligational authority brought forward[A]: $8,677,975.
Fiscal year: 2000;
Amount appropriated: $20,000,000;
Obligational authority brought forward[A]: $10,636,707.
Fiscal year: 2001;
Amount appropriated: $20,000,000;
Obligational authority brought forward[A]: $8,996,504.
Fiscal year: 2002;
Amount appropriated: $20,000,000;
Obligational authority brought forward[A]: $9,729,370.
Fiscal year: 2003;
Amount appropriated: $19,870,000;
Obligational authority brought forward[A]: $2,573,363[B].
Fiscal year: 2004;
Amount appropriated: $20,000,000;
Obligational authority brought forward[A]: $9,394,960.
Fiscal year: 2005;
Amount appropriated: $20,000,000;
Obligational authority brought forward[A]: $18,000,072.
Fiscal year: 2006;
Amount appropriated: $28,000,000;
Obligational authority brought forward[A]: $13,689,240.
Fiscal year: 2007;
Amount appropriated: $31,000,000;
Obligational authority brought forward[A]: $10,465,962.
Source: FEMA's Integrated Financial Management Information System
(IFMIS).
[A] Prior year funds that have not expired and have not been obligated.
[B] According to FEMA, part of the obligational authority brought
forward from fiscal year 2002 to fiscal year 2003 was transferred to
DHS, which was established in fiscal year 2003.
[End of table]
For each of fiscal years 2006 and 2007, Congress appropriated $50
million for the RFC and SRL programs. In fiscal year 2006, Congress
apportioned $10 million specifically for the RFC program and $40
million for the SRL pilot program.[Footnote 37] For fiscal year 2007,
Congress appropriated a sum of $50 million for the two programs, which
FEMA apportioned the same way.[Footnote 38] RFC funds are intended to
assist states and communities in reducing flood damages to insured
properties that have had one or more claims of more than $1,000 to
NFIP. While funding for the SRL pilot program has been made available
since 2006, the application period did not open until January 2008 when
program guidance was published. The purpose of the SRL program is to
provide grant funds for mitigation projects that reduce or eliminate
the long term risk of flood damage to severe repetitive loss
residential structures insured under NFIP.[Footnote 39]
Lack of Current Property-Level Information Limits FEMA's Ability to
Accurately Account for Property Acquisitions:
According to FEMA data, the FMA program acquired 35 percent of the
properties approved for acquisition from fiscal years 1997 through
2007, and the RFC program acquired 19 percent of the properties
approved for acquisition from 2006 through 2007. However, FEMA
headquarters cannot readily account for the actual number of properties
acquired through each of these programs because it does not require
real-time reporting of individual acquisitions, and its grant
management system does not record purchase information. According to
FEMA data, as of October 2007, FEMA approved the acquisition of 980
properties through the FMA program from 1997 through the end of fiscal
year 2007. However, FEMA does not routinely track how many of these
properties have been acquired until a project closes, which can take
several years. According to FEMA's records, 342 of the 980 properties
(35 percent) approved through the FMA program have been acquired
through acquisition-only projects, or mixed mitigation projects, which
involve acquisition and at least one other type of mitigation, such as
relocation, elevation, or floodproofing.[Footnote 40]
Of the 980 properties FEMA approved for acquisition, 952 were approved
as acquisition-only projects. As of October 2007, 331 of the 952
properties approved for acquisition were recorded as acquired. However,
this number is likely understated because FEMA does not account for
property acquisitions until the projects are closed, and no projects
have been recorded as closed since 2003 (see table 5). Based on FEMA
data provided on properties acquired through the FMA program, the
average acquisition amount for acquisition-only projects was more than
$55,000.[Footnote 41]
Table 5: Properties Approved for Acquisition and Acquired through the
FMA Program, 1997-2007 (Acquisition-Only Projects):
Fiscal year approved: 1997;
Approved: Number of projects: 28;
Approved: Total properties in projects: 75;
Closed as of October 2007: Number of projects: 23;
Closed as of October 2007: Total properties in projects: 39.
Fiscal year approved: 1998;
Approved: Number of projects: 27;
Approved: Total properties in projects: 71;
Closed as of October 2007: Number of projects: 27;
Closed as of October 2007: Total properties in projects: 71.
Fiscal year approved: 1999;
Approved: Number of projects: 43;
Approved: Total properties in projects: 116;
Closed as of October 2007: Number of projects: 36;
Closed as of October 2007: Total properties in projects: 72.
Fiscal year approved: 2000;
Approved: Number of projects: 52;
Approved: Total properties in projects: 130;
Closed as of October 2007: Number of projects: 41;
Closed as of October 2007: Total properties in projects: 71.
Fiscal year approved: 2001;
Approved: Number of projects: 72;
Approved: Total properties in projects: 139;
Closed as of October 2007: Number of projects: 44;
Closed as of October 2007: Total properties in projects: 63.
Fiscal year approved: 2002;
Approved: Number of projects: 51;
Approved: Total properties in projects: 83;
Closed as of October 2007: Number of projects: 12;
Closed as of October 2007: Total properties in projects: 13.
Fiscal year approved: 2003;
Approved: Number of projects: 18;
Approved: Total properties in projects: 79;
Closed as of October 2007: Number of projects: 2;
Closed as of October 2007: Total properties in projects: 2.
Fiscal year approved: 2004;
Approved: Number of projects: 14;
Approved: Total properties in projects: 32;
Closed as of October 2007: Number of projects: 0;
Closed as of October 2007: Total properties in projects: 0.
Fiscal year approved: 2005;
Approved: Number of projects: 21;
Approved: Total properties in projects: 49;
Closed as of October 2007: Number of projects: 0;
Closed as of October 2007: Total properties in projects: 0.
Fiscal year approved: 2006;
Approved: Number of projects: 33;
Approved: Total properties in projects: 112;
Closed as of October 2007: Number of projects: 0;
Closed as of October 2007: Total properties in projects: 0.
Fiscal year approved: 2007;
Approved: Number of projects: 18;
Approved: Total properties in projects: 66;
Closed as of October 2007: Number of projects: 0;
Closed as of October 2007: Total properties in projects: 0.
Fiscal year approved: Total;
Approved: Number of projects: 377;
Approved: Total properties in projects: 952;
Closed as of October 2007: Number of projects: 185;
Closed as of October 2007: Total properties in projects: 331.
Source: FEMA.
[End of table]
According to FEMA, the remaining 28 of 980 properties approved through
the FMA program were approved as mixed mitigation projects. Of the 28
properties approved since 1997, 11 were recorded as acquired (see table
6). However, FEMA headquarters cannot reliably account for the actual
number of properties acquired because it does not collect data on
individual acquisitions until a project closes, and no mixed mitigation
projects have been reported as closed since 1999.
Table 6: FMA Program Properties Approved for Acquisition and Acquired,
1997-2007 (Mixed Mitigation Projects):
Fiscal year approved: 1997;
Approved[A]: Number of projects: 1;
Approved[A]: Properties in project: 9;
Approved[A]: Properties to be acquired: 6;
Approved[A]: Properties to be otherwise mitigated[B]: 3;
Closed as of October 2007: Number of projects: 1;
Closed as of October 2007: Properties in project: 9;
Closed as of October 2007: Properties acquired: 6;
Closed as of October 2007: Properties otherwise mitigated[B]: 3.
Fiscal year approved: 1998;
Approved[A]: Number of projects: 2;
Approved[A]: Properties in project: 8;
Approved[A]: Properties to be acquired: 3;
Approved[A]: Properties to be otherwise mitigated[B]: 5;
Closed as of October 2007: Number of projects: 2;
Closed as of October 2007: Properties in project: 8;
Closed as of October 2007: Properties acquired: 3;
Closed as of October 2007: Properties otherwise mitigated[B]: 5.
Fiscal year approved: 1999;
Approved[A]: Number of projects: 1;
Approved[A]: Properties in project: 5;
Approved[A]: Properties to be acquired: 2;
Approved[A]: Properties to be otherwise mitigated[B]: 3;
Closed as of October 2007: Number of projects: 1;
Closed as of October 2007: Properties in project: 5;
Closed as of October 2007: Properties acquired: 2;
Closed as of October 2007: Properties otherwise mitigated[B]: 3.
Fiscal year approved: 2000;
Approved[A]: Number of projects: 0;
Approved[A]: Properties in project: 0;
Approved[A]: Properties to be acquired: 0;
Approved[A]: Properties to be otherwise mitigated[B]: 0;
Closed as of October 2007: Number of projects: 0;
Closed as of October 2007: Properties in project: 0;
Closed as of October 2007: Properties acquired: 0;
Closed as of October 2007: Properties otherwise mitigated[B]: 0.
Fiscal year approved: 2001;
Approved[A]: Number of projects: 1;
Approved[A]: Properties in project: 17;
Approved[A]: Properties to be acquired: 9;
Approved[A]: Properties to be otherwise mitigated[B]: 8;
Closed as of October 2007: Number of projects: 0;
Closed as of October 2007: Properties in project: 0;
Closed as of October 2007: Properties acquired: 0;
Closed as of October 2007: Properties otherwise mitigated[B]: 0.
Fiscal year approved: 2002;
Approved[A]: Number of projects: 2;
Approved[A]: Properties in project: 12;
Approved[A]: Properties to be acquired: 3;
Approved[A]: Properties to be otherwise mitigated[B]: 9;
Closed as of October 2007: Number of projects: 0;
Closed as of October 2007: Properties in project: 0;
Closed as of October 2007: Properties acquired: 0;
Closed as of October 2007: Properties otherwise mitigated[B]: 0.
Fiscal year approved: 2003;
Approved[A]: Number of projects: 0;
Approved[A]: Properties in project: 0;
Approved[A]: Properties to be acquired: 0;
Approved[A]: Properties to be otherwise mitigated[B]: 0;
Closed as of October 2007: Number of projects: 0;
Closed as of October 2007: Properties in project: 0;
Closed as of October 2007: Properties acquired: 0;
Closed as of October 2007: Properties otherwise mitigated[B]: 0.
Fiscal year approved: 2004;
Approved[A]: Number of projects: 0;
Approved[A]: Properties in project: 0;
Approved[A]: Properties to be acquired: 0;
Approved[A]: Properties to be otherwise mitigated[B]: 0;
Closed as of October 2007: Number of projects: 0;
Closed as of October 2007: Properties in project: 0;
Closed as of October 2007: Properties acquired: 0;
Closed as of October 2007: Properties otherwise mitigated[B]: 0.
Fiscal year approved: 2005;
Approved[A]: Number of projects: 1;
Approved[A]: Properties in project: 8;
Approved[A]: Properties to be acquired: 3;
Approved[A]: Properties to be otherwise mitigated[B]: 5;
Closed as of October 2007: Number of projects: 0;
Closed as of October 2007: Properties in project: 0;
Closed as of October 2007: Properties acquired: 0;
Closed as of October 2007: Properties otherwise mitigated[B]: 0.
Fiscal year approved: 2006;
Approved[A]: Number of projects: 1;
Approved[A]: Properties in project: 5;
Approved[A]: Properties to be acquired: 1;
Approved[A]: Properties to be otherwise mitigated[B]: 4;
Closed as of October 2007: Number of projects: 0;
Closed as of October 2007: Properties in project: 0;
Closed as of October 2007: Properties acquired: 0;
Closed as of October 2007: Properties otherwise mitigated[B]: 0.
Fiscal year approved: 2007;
Approved[A]: Number of projects: 1;
Approved[A]: Properties in project: 7;
Approved[A]: Properties to be acquired: 1;
Approved[A]: Properties to be otherwise mitigated[B]: 6;
Closed as of October 2007: Number of projects: 0;
Closed as of October 2007: Properties in project: 0;
Closed as of October 2007: Properties acquired: 0;
Closed as of October 2007: Properties otherwise mitigated[B]: 0.
Fiscal year approved: Total;
Approved[A]: Number of projects: 10;
Approved[A]: Properties in project: 71;
Approved[A]: Properties to be acquired: 28;
Approved[A]: Properties to be otherwise mitigated[B]: 43;
Closed as of October 2007: Number of projects: 4;
Closed as of October 2007: Properties in project: 22;
Closed as of October 2007: Properties acquired: 11;
Closed as of October 2007: Properties otherwise mitigated[B]: 11.
Source: FEMA.
[A] Approved for mitigation.
[B] Properties that were to be mitigated through means other than
acquisition.
[End of table]
FEMA has approved the acquisition of 79 properties through the RFC
program since 2006.[Footnote 42] As of October 2007, FEMA regional
offices confirmed that mitigation activities for 15 properties (19
percent) were completed, although none of these projects have been
closed officially.[Footnote 43]
According to FEMA officials, the data on property acquisitions are
limited because they include only data for completed, or closed,
projects. The completion of a mitigation project, which can include
mitigation activities for multiple properties, may take several years.
The officials stated that a variety of factors can slow or hinder
property acquisitions including historic preservation concerns or
environmental issues. However, FEMA officials stated that the agency
does not require FEMA regional offices to record property acquisitions
until an entire project closes. Internal control standards for the
federal government direct that transactions be promptly recorded to
maintain their relevance and value to management in controlling
operations and making decisions.[Footnote 44] While FEMA guidance for
mitigation programs requires grantees to submit quarterly performance
reports on significant activities, whether work will be completed
within the performance period, and cost information, this guidance does
not explicitly require the reporting of property acquisitions. The lack
of an explicit requirement for FEMA regional offices to record property
acquisitions data in a consistent and timely manner hinders FEMA's
ability to account for the extent to which flood-damaged and repetitive
loss properties have been acquired through its mitigation programs
(thereby decreasing the inventory of these properties).
In addition to lacking a requirement to record acquisition data
promptly, the information system that FEMA currently uses for the NFIP
mitigation programs does not record property acquisition data at all.
Prior to 2004, FEMA regional staff recorded data on properties acquired
through closed projects into FEMA's National Emergency Management
Information System (NEMIS). In 2004, FEMA transitioned from NEMIS to
the Electronic Grants Management System (eGrants) in an effort to meet
the intent of the Electronic Government (eGovernment) initiative.
[Footnote 45] While eGrants was envisioned to help manage all phases of
grants, the system currently assists in the front end of the grant life
cycle by providing grant applicants the ability to electronically
search and apply for grants. The eGrants system does not have the
capability to capture any acquisition data. That is, eGrants records
neither ongoing nor completed property acquisition data for FEMA's
mitigation programs. According to FEMA officials, FEMA is in the
process of consolidating all nondisaster grants into one information
system that will support the entire grants life cycle from solicitation
through closeout.[Footnote 46] The enhanced information system will
track property acquisitions in real-time and will have project closeout
capabilities; however, this information system is not scheduled to be
completed until 2010. Until the nondisaster grants management system is
developed, FEMA will continue to manage grants using eGrants, despite
its limited capabilities.
While FEMA's headquarters officials explained that, since 2004 their
regional staff have used paper files and electronic spreadsheets to
maintain information on property acquisitions for closed projects, FEMA
did not provide information on project closures that have occurred
since 2004 for the FMA program because they could not easily collect
the information from their regional offices. Specifically, when we
requested property acquisition data for the 10-year period of our
review, FEMA was able to provide data on closed projects for 1997
through 2003 (6 of 10 years) for the FMA program, and for 2006, the
first year of the RFC program. FEMA had to collect data from at least
two information systems for the FMA program and from paper files for
the RFC program. GAO has previously reported on the continued lack of
standardization and other inefficiencies in grant administration across
agencies, including not aligning federal grant processes with typical
grantees' business processes, inadequate advance information on
potential grant availability, and unexplained delays in grant awards.
[Footnote 47] However, a lack of standardization does not preclude FEMA
from having a mechanism that allows it to effectively track real-time
property acquisition data.
As a result of the limitations of eGrants in capturing property
acquisition data, FEMA cannot readily determine the extent to which
NFIP-funded mitigation programs have been reducing or eliminating the
risk of losses from repetitive loss properties, a key measure needed to
determine the effectiveness of the program. Coupled with lack of a
requirement for staff to consistently and promptly record acquisitions
data, the lack of comprehensive data impedes FEMA's ability to account
for the properties acquired through its mitigation programs and hinders
its ability to provide Congress with timely information with which to
assess the impact of these programs on mitigating losses to NFIP and
ensure that limited resources are appropriately allocated.
FEMA's Oversight of Contractor Activities Lacked Consistency,
Documentation, and Coordination:
FEMA makes extensive use of contractors who perform critical functions
to implement the NFIP. Contractors perform functions such as the
collection and reporting of all financial and statistical data and the
selling and servicing of some flood insurance policies. While FEMA
relies upon contractors to implement the NFIP, our review of FEMA's
monitoring documentation for the BSA contract showed that FEMA did not
consistently follow its own monitoring procedures for preparing,
maintaining, and reviewing monitoring reports, and was unable to
provide copies of the majority of the monitoring reports we requested.
Moreover, key FEMA offices that have responsibilities for addressing
contractor deficiencies did not coordinate information and actions
related to deficiencies identified for the BSA and DSA contractors. As
a result, FEMA could not ensure adherence to contract requirements and
lacked information critical for effective oversight on key NFIP
contractors.
FEMA Staff Responsible for Regularly Monitoring BSA Performance Areas
Did Not Clearly or Consistently Follow Monitoring Procedures:
Our review found that FEMA staff did not consistently follow monitoring
procedures related to the submission, documentation, and review of
monitoring reports for the BSA contract. In assessing the extent to
which FEMA staff followed their contract-specific procedures, we
focused on monitoring of the BSA contractor, which is responsible for
all financial and statistical reporting related to NFIP.
FEMA lacked comprehensive monitoring policies and procedures for all
NFIP-related contracts during the period of our review; instead, FEMA's
procedures were summarized in individual contracts, and, for the BSA
contract we reviewed, in a "surveillance plan" that specified
monitoring roles and responsibilities. According to the BSA contract,
in effect from October 2005 through December 2007, FEMA monitors were
to review the BSA's performance in 10 specific performance areas. FEMA
assigned five monitors including a COTR to the BSA contract, and each
was required to prepare monitoring reports containing the results of
their specific surveillance activities.[Footnote 48] The reports were
to indicate clearly the performance area monitored and whether the
performance standard was met, according to an official from the Program
Management Office. The surveillance plan also specified that reports
were to be submitted in a timely fashion to the COTR and maintained by
the COTR for the life of the contract in a quality assurance file. Both
the COTR and the Program Management Office were supposed to review the
monitoring reports.
In March 2008, FEMA provided GAO with a set of contract monitoring
procedures that are applicable to all Risk-Insurance NFIP contracts
that have performance requirements. However, these procedures were not
scheduled to be implemented until May 1, 2008.
FEMA Lacks Records for Most Monitoring Reports We Requested, and
Available Reports Were Not Clearly Linked to Performance Requirements:
FEMA did not clearly or consistently follow the requirements to submit
a monitoring report as specified in both the BSA contract and the
surveillance plan. Seven of these performance areas were to be
monitored monthly, one was to be monitored quarterly, and two were to
be monitored annually. For the time period we reviewed (from October
2005 through May 2007), FEMA monitors should have produced 380 monthly
monitoring reports. In response to our request for all of these
monitoring reports, FEMA made 145 available for our review. While
monitors stated that some of the missing reports were submitted to
either the COTR or the Program Management Office, the monitor, COTR,
and the Program Management Office were not able to produce evidence of
these reports or their findings. FEMA officials stated that many
reports were not prepared because most monitors were detailed to assist
with recovery efforts following Hurricane Katrina. However, our
analysis shows the majority of the monthly monitoring reports that were
required to be prepared throughout 2006 apparently were not.
Further, more than 70 percent of the 145 monthly reports we reviewed
did not specify the performance areas to which they related (see table
7).[Footnote 49] For example, some monitoring reports described the
contractor's activities, without specifying what performance area was
being reviewed or an assessment of the performance. Because some
monitors were responsible for reporting on multiple performance areas,
and did not specify the areas on which they were reporting, we could
not consistently determine the performance area(s) that the reports
were intended to address.
Table 7: Extent to Which FEMA Produced Required Monthly Monitoring
Reports for the BSA Contractor (October 2005-May 2007):
Performance area: 1 - Statement of Work;
Monitoring reports required by contract: 100;
Monitoring reports provided to GAO: 42;
Monitoring reports that GAO could link to a performance area: 20.
Performance area: 3 - Implementation of Program Management Plans;
Monitoring reports required by contract: 100;
Monitoring reports provided to GAO: 42;
Monitoring reports that GAO could link to a performance area: 4.
Performance area: 5 - Disaster Response;
Monitoring reports required by contract: 20;
Monitoring reports provided to GAO: 12;
Monitoring reports that GAO could link to a performance area: 12.
Performance area: 6 - Program Development, Operations, and Systems;
Monitoring reports required by contract: 100;
Monitoring reports provided to GAO: 36;
Monitoring reports that GAO could link to a performance area: 1.
Performance area: 7 - Timeliness of Service;
Monitoring reports required by contract: 20;
Monitoring reports provided to GAO: 4;
Monitoring reports that GAO could link to a performance area: 1.
Performance area: 8 - Quality of Customer Service;
Monitoring reports required by contract: 20;
Monitoring reports provided to GAO: 4;
Monitoring reports that GAO could link to a performance area: 0.
Performance area: 10 - Timeliness of Prior Term Refunds;
Monitoring reports required by contract: 20;
Monitoring reports provided to GAO: 5;
Monitoring reports that GAO could link to a performance area: 0.
Performance area: Total;
Monitoring reports required by contract: 380;
Monitoring reports provided to GAO: 145;
Monitoring reports that GAO could link to a performance area: 38.
Source: GAO analysis of FEMA's monitoring reports for the BSA contract.
Notes: This table focuses on monthly monitoring reports for the BSA
contract. Thus, it excludes analysis of performance area 9, which was
to be monitored quarterly, and performance areas 2 and 4, which were to
be monitored annually. We arrived at the number of "reports required by
contract" by multiplying the number of monitors who were required to
monitor the performance areas with the number of months for which GAO
requested information.
[End of table]
Many of the monthly reports were not clearly linked to contract-
specified requirements because they used performance area names and
numbers that were inconsistent with names and numbers in the BSA
contract. When asked about the lack of a direct link to specific
performance areas, as listed in the BSA contract, most of the BSA
monitors we interviewed stated that they did not know the "number" of
the performance area they were to monitor but knew the general areas.
When we brought this issue to the attention of FEMA's Program
Management Office in August 2007, it determined that the numbering of
performance areas was inconsistent in the BSA contract and the
surveillance plan. As a result, the office revised the surveillance
plan to make numbering of performance areas consistent with the
contract. The surveillance plan also was modified to include the
specific names of the monitors responsible for each performance area.
FEMA's inability to provide documentation that clearly indicated how
contractor performance linked to contract standards indicates that
records had not been properly managed and maintained in accordance with
internal control standards for the federal government.[Footnote 50]
FEMA management acknowledged the problem when we presented a
preliminary analysis of performance reports and performance standards.
While we did not become aware of any significant performance problems
with the BSA contractor during the course of our review, our results
are inconclusive because most of the monitoring reports were not
documented. Given FEMA's extensive use of contractor's to implement the
NFIP, it is vital that FEMA maintain internal controls, which are
critical to ensuring that contractors are performing as required. The
lack of documentation and linkages between performance reports and
performance areas and the extent to which FEMA effectively monitored
the BSA contract revealed weaknesses that could adversely impact the
functioning of the NFIP.
For Most of Available Reports, We Found Little Evidence of Required
Timely Submission or Systematic Review and Maintenance of
Documentation:
The majority of the 145 monthly monitoring reports that we reviewed
were not date stamped or otherwise annotated to reflect the date they
were submitted to either the COTR or the Program Management Office.
Moreover, an official from the Program Management Office was unable to
provide evidence that her office tracked the dates on which monitoring
reports were received from each monitor.
In response to our and other queries about the timeliness with which
monitoring reports were submitted to the Program Management Office, in
July 2007, the office directed monitors to submit their monthly
monitoring reports to the COTR on the fifth of each month. The COTR was
then directed to review these reports and provide the Program
Management Office with a summary of the findings by the tenth of each
month. While we did not evaluate monitoring reports submitted after May
2007, these contract-specific policies are a step in the right
direction for ensuring that monitoring reports are collected and
reviewed in a timely manner.
In addition, the COTR and the Program Management Office appeared not to
have consistently reviewed the monitoring reports. For example,
according to the COTR, the Program Management Office directed monitors
to send their monitoring reports directly to the Program Management
Office. The COTR did not begin to receive them until July 2007, 21
months into the 27-month BSA contract. Therefore, it was not until July
2007 that the COTR began maintaining a quality assurance file for the
BSA contract. While FEMA provided documentation of a portion of the
required monitoring reports, FEMA could not provide evidence that they
were reviewed systematically by the COTR or the Program Management
Office. Internal control standards for the federal government provide
that ongoing monitoring that occurs in the normal course of operations
is performed continually and is ingrained in the agency's operations.
[Footnote 51] Moreover, these standards call for clear areas of
authority and appropriate lines of reporting. While the surveillance
plan established that the COTR had responsibility for collecting
monitoring reports from each monitor and maintaining them in a quality
assurance file, he was not a part of the process for reviewing
monitoring reports until July 2007.
As a result of these conditions, the extent to which monitoring reports
were being reviewed appropriately and in a timely fashion was unclear.
Such conditions could allow performance problems to go unnoticed and
potentially worsen, thereby affecting the quality of NFIP. In addition,
the failure of the COTR to begin maintaining a quality assurance file
until July 2007 meant that a complete file of monitoring reports could
not be provided to the Contracting Officer for consideration during
contract renewal negotiations. Although the BSA contract was awarded to
a different contractor in January 2008, the Contracting Officer would
have been unable to determine whether the original contractor had a
record of performance problems, potentially useful information to
consider when deciding whether to renew a contract.
In March 2008, FEMA provided GAO with monitoring procedures that
established a clear process through which monitoring reports should be
submitted to the COTR and the Program Management Office, and the review
steps that should be taken by the COTR. However, these procedures do
not specify the role of the Program Management Office in reviewing
monitoring reports, or its broader responsibilities for the
implementation of the contract monitoring procedures. Again, such
conditions could allow performance problems to go unnoticed and
potentially worsen, thereby affecting the quality of NFIP.
FEMA's Handling of BSA and DSA Contractor Deficiencies Was Not
Coordinated and Was Poorly Documented:
Analysis of FEMA's efforts to address performance deficiencies for two
major NFIP contractors--the BSA and DSA--indicates a lack of
coordination between key FEMA officials and a failure to properly
document decisions and actions associated with performance problems. Of
the 38 monitoring reports that FEMA provided in relation to the BSA
contract, and that could be linked to specific performance areas, 2
indicated that performance standards were not met. While no discrepancy
reports (which FEMA requires the Contracting Officer to issue when
contractors fail to meet specified performance standards) were written
as a result of the BSA deficiencies, 18 discrepancy reports were issued
for the DSA contract.
While Monitoring Reports Noted Some BSA Deficiencies, FEMA Staff Did
Not Issue Discrepancy Reports or Inform the Contracting Officer of
Deficiencies:
Two of the 38 BSA monthly monitoring reports that could be clearly
linked to performance standards indicated that a performance standard
was not met (see table 8). In both cases, one of which involved the
failure of the BSA's computer security system to meet performance
standards, and the other related to the contractor's failure to submit
a deliverable to FEMA on time, the COTR determined that a discrepancy
report was not necessary and did not inform the Contracting Officer of
the contractor's failure to meet its performance standards. According
to the COTR, a discrepancy report was not necessary in the first case
because the computer security standards were relatively new at the time
and have been updated since. The COTR stated that, in the second case,
he used his discretion to determine that a discrepancy report was not
necessary and resolved the issue through discussions with the
contractor.
Table 8: Analysis of FEMA Monthly Monitoring Reports Required for the
BSA Contract (October 2005-May 2007):
Performance area: 1 - Statement of Work;
Performance requirement was met: 19;
Performance requirement was not met: 1.
Performance area: 3 - Implementation of Program Management Plans;
Performance requirement was met: 4;
Performance requirement was not met: 0.
Performance area: 5 - Disaster Response;
Performance requirement was met: 12;
Performance requirement was not met: 0.
Performance area: 6 - Program Development, Operations and Systems;
Performance requirement was met: 0;
Performance requirement was not met: 1.
Performance area: 7 - Timeliness of Service;
Performance requirement was met: 1;
Performance requirement was not met: 0.
Performance area: 8 - Quality of Customer Service;
Performance requirement was met: 0;
Performance requirement was not met: 0.
Performance area: 10 - Timeliness of Prior Term Refunds;
Performance requirement was met: 0;
Performance requirement was not met: 0.
Performance area: Total;
Performance requirement was met: 36;
Performance requirement was not met: 2.
Source: GAO analysis of FEMA's monitoring reports for the BSA contract.
Note: This table focuses on monthly monitoring reports for the BSA
contract. Thus, it excludes analysis of performance area 9, which was
to be monitored quarterly, and performance areas 2 and 4, which were to
be monitored annually.
[End of table]
While only a few deficiencies were identified, the monthly monitoring
reports that we reviewed most often found that the performance
requirement was met. In two instances, they revealed a lack of
coordination between the BSA COTR and the Contracting Officer. The BSA
contract stated that if a performance standard was not met, the
Contracting Officer was to issue a discrepancy report to which the
contractor had to respond. Upon evaluation of the contractor's
response, the Contracting Officer would determine whether a deduction
in payment to the contractor was appropriate. The failure of the COTR
to share his assessment may be due to the lack of written policies and
procedures specifying how such assessments should be shared with the
Contracting Officer. While FEMA developed guidance in July 2007
specifying when and how the COTR should communicate discrepancies with
the Contracting Officer, this guidance is specifically for the BSA
contract. Federal standards for internal control call for appropriate
documentation of significant events. Because the COTR did not document
and inform the Contracting Officer of his assessment, FEMA management
was unaware of the contractor's failure to meet a performance
requirement; the contractor was not penalized financially.[Footnote 52]
FEMA's Handling of Discrepancy Reports for the DSA Contract Was
Uncoordinated and Hampered by Lack of Guidance:
Our analysis of the actions taken in connection with the 18 discrepancy
reports written from September 2004 through May 2007 on the DSA
contract also found that FEMA offices did not coordinate in preparing,
documenting, and reviewing these determinations. Specifically, the COTR
for the DSA contract, FEMA's Program Management Office, and FEMA's
Contracting Officer did not coordinate to address the failure of the
DSA to meet contract-specified performance standards.[Footnote 53]
Generally, the DSA received discrepancy reports for not processing NFIP
policy documents, such as renewals and applications, within required
time frames. According to the COTR, these types of deficiencies had
minimal negative effects because, in some cases, the universe of
documents the contractor processed was so small that the failure to
process 10-20 documents in a timely manner resulted in a discrepancy
report. But, interviews with the FEMA offices responsible for
processing discrepancy reports revealed that FEMA's Contracting Officer
was unaware that the COTR for that contract had sent 18 discrepancy
reports to the DSA. According to the Contracting Officer, her office
became aware that discrepancy reports had been sent to the DSA in June
2007 during the course of our review.
In addition to a lack of communication about the discrepancy reports,
some reports lacked required signatures, or evidence that different
offices had reviewed the contractor's responses and determined what, if
any, actions would be taken. The discrepancy report is to contain a
signature from FEMA's Contracting Officer; a response from the
contractor; FEMA's subsequent evaluation of the contractor's response;
and the resulting action, such as a payment reduction. While
discrepancy reports were to be signed by the Contracting Officer, none
of the reports we reviewed had the signatures. The lack of signatures
by the Contracting Officer suggests that this officer was unaware of
the performance problems associated with the DSA and potentially could
have renewed the contract without complete information about the
contractor's performance.
Our review also showed a failure to consider the contractor's written
response to the report. According to the DSA contract, FEMA's
determination of payment changes is to be based on consideration of the
contractor's response to the discrepancy report. Neither the contract
nor the discrepancy reports specify how the Contracting Officer was to
be made aware of contractor performance problems; that is, whether and
how either the COTR or the Program Management Office was responsible
for informing the Contracting Officer of contractor performance
problems. Six reports contained a written response from the contractor,
yet none of the reports contained information on FEMA's assessment of
the contractor's response or its subsequent actions (for example,
payment reductions). According to the COTR for the DSA contract, if a
contractor did not meet a performance standard, the COTR would arrange
for deductions, regardless of the contractor's stated or written
response.
Moreover, we found that the 18 discrepancy reports called for a total
of more than $55,000 in financial disincentives for the DSA. However,
FEMA had not debited nearly $19,000 in financial disincentives from the
contractor's invoices as of January 2008. According to the COTR, the
disincentives that had not been debited had been assessed in 6
discrepancy reports prepared from 2005 through 2006. The COTR never
gave 2 of these discrepancy reports to FEMA's accounting department for
processing, and the accounting department did not process the remaining
4. According to the COTR, all 6 discrepancy reports were not properly
processed due to human error. The COTR said that FEMA had no process by
which he could be informed that a financial penalty had been processed
as a result of the discrepancy reports. In addition to the 6
unprocessed reports that assessed disincentives, FEMA did not process 2
other reports because they were drafted within the first 6 months of
the contract. According to the COTR, the contractor had a 6-month
"grace period" for meeting its performance standards; therefore, no
financial disincentives were either calculated or applied for these
discrepancy reports. However, such a grace period is not specified in
the DSA contract. The lack of clear, written policies and procedures
regarding the specific roles of various FEMA offices in processing
discrepancy reports and ensuring the appropriate application of
financial disincentives resulted in the failure of FEMA to process
financial disincentives.
In the monitoring procedures that FEMA provided to GAO in March 2008,
the role of the COTR in identifying deficiencies and communicating them
with a representative of the Contracting Officer is specified. However,
the procedure does not explain how the Program Management Office will
be involved, or how the agency will ensure that financial disincentives
are appropriately applied.
Conclusions:
Various statistical trends reveal that while NFIP is growing and
becoming more widely purchased enabling it to more widely diversify its
risks, FEMA continues to face a number of challenges in managing this
program including a growing number of claims on repetitive loss
properties, insufficient information to readily track the property
acquisitions of its mitigation programs, inconsistent application of
contract monitoring procedures, and inconsistent documentation and
coordination of contractor oversight. The ability of FEMA officials to
effectively monitor the effectiveness of its flood mitigation programs,
as well as its NFIP contractors, is essential to the effective
implementation of NFIP.
Floods impose an enormous financial burden on the nation's flood
insurer, the federal government. Consequently, FEMA management,
Congress, and the public need accurate and timely information to assess
the effectiveness of programs designed to mitigate flood-related
damages, particularly for repetitive loss properties that account for a
significant portion of claims paid under NFIP. We found that FEMA does
not have specific written policies for the timely recording of property
acquisition data in its mitigation programs. By developing such
policies and guidance, FEMA could improve its ability to produce,
analyze, and report on property acquisitions and thus improve its
ability to assess the operations and effectiveness of these programs.
In addition to not requiring the timely recording of acquisition data,
since 2004 when FEMA began using eGrants as its grants management
system, FEMA has been unable to use this grants management system to
record such information. The eGrants system can process "front-end"
processes (such as accepting applications) but not subsequent grants
management processes (such as project closures). Not having an end-to-
end grant management capability makes it particularly challenging for
grant management staff to oversee the grant process across their
organization. Grant management systems were intended to enhance the
ability of federal managers and grant-making agencies to readily
collect, analyze, and report information. With its current system
limitations, FEMA cannot accurately assess or report the extent to
which its mitigation programs are reducing the number of flood-damaged
and repetitive loss properties.
Deficiencies in FEMA processes relating to NFIP data and performance
extend to oversight of contractors responsible for supporting NFIP. For
instance, we were unable to link a substantial number of monitoring
reports to contract-specified performance areas and were unable to
determine, based upon available documentation, the extent to which
monthly monitoring reports were submitted on time, systematically
reviewed by appropriate officials and offices, and the associated
documentation properly filed and maintained. The lack of a framework to
ensure that essential oversight and control functions were completed
could allow performance problems to go unnoticed or worsen.
Furthermore, the lack of documentation to determine the extent to which
specific performance standards were met makes it appear that FEMA was
not effectively monitoring its contractors, diminishes its ability to
determine if and when problems occurred, and may have led the agency to
not apply specific financial incentives or disincentives per contract
terms (such as when the BSA COTR independently determined that
discrepancy reports were not necessary). Given the reliance of NFIP
upon contractors, poor contractor performance could diminish the
overall quality of program operations and management.
In addition to enhancing its management and oversight of its
contractors, FEMA also has an opportunity to improve internal
operations and communication as they relate to contract oversight
functions. Our review indicated a lack of overall contract-monitoring
guidance and consequently conflicting understandings of procedures,
responsibilities, and authority for addressing contractor deficiencies
among FEMA officials and offices. While FEMA developed guidance during
the course of our review for all Risk-Insurance NFIP-related contracts
that have performance requirements, in part as a result of issues that
we identified during our work, revised monitoring procedures were not
scheduled to be implemented until May 2008, which was after the period
of our analysis.
Recommendations for Executive Action:
To more accurately track the extent to which flood-prone properties are
acquired, we recommend that the Secretary of Homeland Security take the
following two actions:
* Establish written guidance for FEMA regional offices to better ensure
consistent and timely recording of property acquisition data.
* Establish a means to track real-time property acquisitions for NFIP-
funded mitigation programs.
To ensure more effective oversight of contractors performing key NFIP
data collection, reporting, and insurance functions, we also recommend
that the Secretary of Homeland Security take the following three
actions:
* Implement a process to ensure that monitoring reports are submitted
on time and systematically reviewed by the COTR and the Program
Management Office and copies of monitoring reports are retained in a
quality assurance file, as directed by the contract.
* Ensure that FEMA staff clearly monitor each performance standard that
the contractor is required to meet in the time frames required by
contract and that FEMA staff clearly link monitoring reports and
performance areas.
* Ensure implementation of written guidance for all NFIP-related
contracts on how to consistently handle the failure of a contractor to
meet standards in performance areas and establish written policies and
procedures about the coordination between FEMA officials and offices
(including the COTR, the Program Management Office, and the Contracting
Officer) when addressing contractor deficiencies, including determining
whether and under what circumstances to issue discrepancy reports, and
ensuring that financial disincentives are appropriately and
consistently applied.
Agency Comments and Our Evaluation:
We requested comments on the draft of this report from the Secretary of
Homeland Security. The Assistant Administrator, Mitigation Directorate,
FEMA, provided written comments that are reprinted in appendix VIII.
FEMA's written response also included technical comments, which we
incorporated as appropriate. In its written comments, FEMA generally
concurred with our recommendations and noted that the agency has taken
actions to address the recommendations related to oversight of
contractors performing key data collection, reporting, and insurance
functions. In particular, the agency stated that it has:
* implemented a procedure whereby monitors will provide a monthly
status report to the COTR using a specific format to ensure monitoring
is performed in a consistent manner across all contracts; and:
* developed written procedures on monitoring reporting requirements and
conducted a training session with all monitors.
In commenting on our presentation of information related to mitigation
programs that can be used to acquire properties, FEMA headquarters
agreed that the agency does not track property acquisition data in real
time; that is, between review of project applications and closeout.
FEMA stated that it would be inefficient for headquarters to collect
real-time data from regional, state, and local partners, and asserted
that the added value of such data was unclear.
We agree that FEMA lacks an efficient process for collecting data on
the extent to which properties have been acquired before a project is
closed out and, therefore, recommend that FEMA establish written
guidance for FEMA regional offices to better ensure consistent and
timely recording of property acquisition data. We disagree that the
value of collecting real-time property acquisition data at the
headquarters level is unclear. Real-time data would improve FEMA
headquarters' ability to produce, analyze, and report information on
ongoing operations and thus improve its ability to assess the
effectiveness of its mitigation programs. For example, FEMA
headquarters would be able to more accurately assess the rate at which
properties are acquired and thus pinpoint within a shorter time frame
the extent to which mitigation programs were reducing the number of
flood-damaged and repetitive loss properties.
As agreed with your office, unless you publicly announce the contents
of this report earlier, we plan no further distribution of this report
until 30 days from the report date. At that time, we will provide
copies to the Chairman, Senate Banking, Housing and Urban Affairs
Committee; the Chairman and Ranking Member of the Senate Committee on
Homeland Security and Governmental Affairs; the Chairman and Ranking
Member of the House Committee on Financial Services; the Chairman and
Ranking Member of the House Committee on Homeland Security; and other
interested committees. We are also sending a copy of this report to the
Secretary of Homeland Security and other interested parties. We also
will make copies available to others upon request. In addition, the
report will available at no charge on our Web site at [hyperlink,
http://www.gao.gov]. Contact points for our Offices of Congressional
Relations and Public Affairs may be found on the last page of this
report.
If you or your staff has any questions about this report, please
contact me at (202) 512-8678 or williamso@gao.gov. GAO contact and
staff acknowledgments are listed in appendix IX.
Sincerely yours,
Signed by:
Orice M. Williams:
Director, Financial Markets and Community Investment:
[End of section]
Appendix I: Objectives, Scope, and Methodology:
Our objectives were to (1) describe trends for National Flood Insurance
Program (NFIP) policies, insurance amounts, premiums, and losses from
1997 through 2006 and the extent to which NFIP losses were attributable
to hurricanes and repetitive loss properties; (2) assess how the
amounts available for the purchase of flood-damaged properties changed
over time, and the extent to which the Federal Emergency Management
Agency (FEMA) purchased flood-damaged properties; and (3) evaluate the
extent to which FEMA followed its procedures for monitoring selected
NFIP-related contracts.
Description of NFIP Trends:
To describe overall NFIP statistics, we obtained data from FEMA on
policies in force and claims made. We obtained data as of December 31
for each year of the period from calendar year 1997 through calendar
year 2006 on policies in force, average insurance per policy in force,
and average premium per policy in force. For the claims data, we
obtained information (as of May 31, 2007) about the claim dates of
losses incurred between January 1, 1997, and December 31, 2006. For all
monetary data obtained, we adjusted dollar amounts to 2006 constant
dollars using the Shelter Total formulas found in the Consumer Price
Index.[Footnote 54]
To describe statistics by flood zone, we first determined in which
flood zones the majority of flood insurance policies were located by
analyzing the data using SAS software version 9.1. FEMA has mapped
flood risks in the United States and in most flood-prone areas has
assigned a flood zone designation based on the level of flood risk.
Although FEMA uses 12 types of flood zones (see appendix VII), we
combined these into four flood zone designations based on risk levels-
-moderate-to low-risk (Zones B, C, and X); high-risk (Zone A); high-
risk coastal (Zone V); and undetermined risk (Zone D). FEMA officials
agreed with the way these were combined. In appendix III we also
present data on Zone "O" (Other), which is not an official flood zone;
rather, FEMA's contractor uses the designation for coding purposes to
indicate missing values or incorrect codes. Less than 1 percent of all
policies and less than .5 percent of all losses fell under Zone "O." In
addition, in appendix III we present data on Zone "E" (Emergency
Program), which indicates policies in the Emergency Program, which does
not assign flood-risk zone designations. Communities participating in
NFIP initially enter the Emergency Program. We determined that, for the
10-year period of our review, 99 percent of policies in force were
located in the high-risk, high-risk coastal, and moderate-to low-risk
zones. We, therefore, reported on these in the body of the report, and
provided data on the remaining zones (undetermined risk and Emergency
Program) in appendix III. For all flood zones, we obtained data on
policies in force, annual average amount of flood insurance purchased,
annual average premium, and losses.
To describe statistics by occupancy type, we first determined which
occupancy types represented the majority of policies in force. We
determined that, for the period of our review, residential (single-
family and two-four family) properties generally accounted for more
than 70 percent of policies in force each year, and condominiums
accounted for more than 20 percent of policies in force each year. We
reported information on these occupancy types in the body of the report
and provided data on the remaining occupancy types (other residential
and nonresidential) in appendix IV.
We conducted electronic testing of FEMA's data to identify outliers and
missing data elements. We also cross-checked various tables to assess
the consistency of the data provided. For example, we compared data on
overall losses with the losses reported by flood zone to ensure that
the losses reported each year were consistent. According to our
electronic testing and cross-checking, we determined that these data
were sufficiently reliable for the purposes of this report.
We obtained data on FEMA's average historical loss estimate (the
purpose of which is to estimate the amount of premium that would be
sufficient to pay for the average level of losses that occurred in past
years) for calendar years 1997 through 2006 from FEMA's Chief Actuary.
We interviewed FEMA officials about the basis for the average
historical loss estimate, reviewed FEMA's white paper on the estimate,
and reviewed FEMA's 2006 Actuarial Rate Review for additional
information about this estimate. We also obtained data on FEMA's
estimates of catastrophic loss years from FEMA's Chief Actuary. FEMA
prepares the catastrophic loss year estimate, usually every other year,
to indicate the maximum amount that NFIP would have to expend during a
catastrophic loss year. That is, the estimate is designed to provide
Congress with an informal guide on the losses that could occur in the
event of a storm that has a 0.1 percent chance of occurring. We did not
assess FEMA's methodology for either of these estimates.
To evaluate the extent to which flood-related damages exceeded flood
insurance limits, we requested and obtained data from FEMA on flood-
insurance policies that were purchased for coverage at the maximum
limit for building-only coverage and contents-only coverage for
calendar years 1997 through 2005. For policies that contained both
building and contents coverage, we obtained data on policies where the
maximum insurance limit was purchased for either the building or
contents coverage (or both) for calendar years 1997 through 2005. We
excluded group flood insurance policies because they have coverage
limits that differ from other flood insurance policies.[Footnote 55] We
obtained data on the number of claims that were paid for the resulting
subset of flood-insurance policies and the number of claims for which
damage data were available. We then identified the number of claims for
which flood-related damages exceeded the maximum amount of insurance
available and the losses associated with these claims. We presented the
data for 1997-2004 and 2005 separately due to the high number of claims
that occurred in 2005. For more information, see appendix VI. In
conjunction with this work, we also reviewed parts of the 2006 study
commissioned by FEMA.[Footnote 56]
To evaluate the extent to which flood-related damages were greater than
the amount of insurance purchased, we obtained data on flood insurance
policies that were purchased for coverage below the maximum limit for
contents coverage, and below the maximum limit for building coverage,
for calendar years 1997 through 2005. For policies that contained both
building and contents coverage, we obtained data on policies where less
than the maximum insurance limit was purchased for either the building
or contents coverage (or both) for calendar years 1997 through 2005. We
excluded group flood insurance policies because they have coverage
limits that differ from other flood insurance policies. We obtained
data on the number of claims that were paid for the resulting subset of
flood insurance policies and the number of claims for which damage data
were available. We then identified the number of claims for which flood-
related damages exceeded the amount of insurance purchased and the
losses associated with these claims. We presented the data for 1997-
2004 and 2005 separately due to the high number of claims that occurred
in 2005.
To describe the extent to which losses were attributable to repetitive
loss properties, we obtained data from FEMA on the net cumulative
number of claims filed for repetitive loss properties and the amount of
losses FEMA paid out for these properties as of December 31 for each
calendar year from 1997 through 2006. We obtained net cumulative data
because this is the only way that FEMA collects data on repetitive loss
properties. We also obtained data on the net cumulative number of
repetitive loss properties for calendar years 1997-2006. According to
FEMA officials, the number of repetitive loss properties represents a
snapshot in time of the number of insured repetitive loss properties as
of the end of each calendar year. The number of claims and amount of
losses paid out on repetitive loss properties represents the net
cumulative total of all dollars paid out. As a result, based upon the
configuration of the available data, we were unable to calculate the
average loss paid out to repetitive loss properties.
To describe the extent to which losses were paid out for hurricane-
damaged properties, we requested information on hurricane-related
claims from FEMA for calendar years 1997 through 2006. According to
FEMA officials, claims are not coded or notated as being hurricane-
related. However, FEMA does maintain a list of "significant events." To
provide GAO data on hurricane-related claims, FEMA identified claims
that were associated with significant events, then identified those
events that were hurricanes. We did not verify FEMA's categorizing of
claims as being "hurricane-related."
Analysis of Amounts Available for Property Acquisition and Extent of
Property Acquisition:
To assess how the amounts available for the purchase of flood-damaged
properties have changed over time, we obtained data from FEMA on
funding appropriations, obligation authority brought forward, and
funding available for the three NFIP-funded mitigation programs.
[Footnote 57] For the Flood Mitigation Assistance (FMA) program, we
obtained data for fiscal years 1997 through 2007. For the Repetitive
Flood Claims (RFC) program and the Severe Repetitive Flood (SRL) pilot
program, we obtained data for fiscal years 2006 and 2007, as both
programs were unfunded until 2006. To assess the extent to which flood-
damaged properties have been acquired, we also obtained data from FEMA
on the number of projects and properties approved for acquisition and
the number of properties actually acquired through the FMA and RFC
programs, but not the SRL program because guidance for this program was
not published until January 2008. FEMA officials informed us that the
most reliable information that they could provide on property
acquisitions was at the project level. FEMA officials also stated that
because not all properties approved for acquisition are ultimately
acquired, it may take several years for a project to close. Because
FEMA's grant management system does not track project closures or
property acquisitions in real time, the data we report is likely an
undercount of the actual number of properties that have been acquired
through the FMA and RFC mitigation programs.
For the FMA program, FEMA officials obtained data on property
acquisitions (for closed projects) that occurred from 1997 through 2003
from FEMA's Enterprise Data Warehouse. According to FEMA officials, the
warehouse data are extracted from the National Emergency Management
Information System (NEMIS), the information system FEMA used to manage
the FMA program until 2003. In 2004, FEMA transitioned from NEMIS to
the Electronic Grants Management System (eGrants) to manage the FMA
program. eGrants does not have project closure capabilities; therefore,
FEMA was unable to provide any data on property acquisitions for FMA
projects that may have closed from 2004 through 2007. While we did not
attempt to verify the reliability of this information, we did look for
inconsistencies between the number of properties approved for
acquisition and the number actually acquired and followed up with FEMA
officials for an explanation of any differences.
The property acquisition data FEMA provided for the FMA program
differentiated acquisition-only projects, and mixed mitigation
projects, or projects that mitigate flood-prone properties through
acquisition and additional mitigation strategies, such as property
elevation. We did not attempt to verify the categorization of
acquisition-only or mixed mitigation projects but presented them as
provided by FEMA.
For the RFC program, which started in 2006, FEMA provided data on
properties acquired (through closed projects) from regional FEMA
offices, which keep such data in electronic spreadsheets or paper
files. We did not attempt to verify the accuracy of this information.
Analysis of Extent to Which FEMA Followed Monitoring Procedures for
Selected NFIP Contracts:
To obtain information on the procedures FEMA uses to monitor its NFIP-
related contracts, we reviewed contracts awarded in support of NFIP,
federal acquisition regulations, a surveillance plan, performance
standards, and the template for discrepancy reports (which FEMA
requires a Contracting Officer to issue when a contractor fails to meet
its performance standards). For selected contracts, we interviewed FEMA
officials and staff with responsibility for issuing and overseeing
contracts--the Contracting Officer, the manager of FEMA's Program
Management Office, the Contracting Officer's Technical Representatives
(COTR) for each contract we reviewed, and monitors--as well as
contractors.
To determine how effectively FEMA followed its monthly monitoring
procedures, we focused upon the contract for FEMA's Bureau and
Statistical Agent (BSA), which is responsible for collecting and
reporting on financial and statistical data for the NFIP. First, we
collected documentation of monthly monitoring reports for the period
from October 2005 through May 2007 to determine if the reports were
submitted by monitors to the appropriate office in a timely manner. We
then compared these reports with the contract-specified performance
standards to see if FEMA reported on each standard and to assess the
extent to which FEMA determined whether the contractor met performance
standards. We asked FEMA officials to provide an explanation of missing
reports and explain why discrepancy reports were not required when
performance standards were not met. In August 2007, we were provided
with a Contract Administration Plan for the BSA contract, dated July
25, 2007 (nearly 22 months after the contract award date) that
documented the contract administration process.
During the course of our work, we learned the NFIP's Direct Servicing
Agent (DSA) contractor had received multiple discrepancy reports from
FEMA. To determine if FEMA had followed its stated policies and
procedures for addressing deficiencies identified through monitoring,
we obtained copies of the discrepancy reports written for the DSA
contractor, documentation of the contractor's responses to the reports,
and documentation of financial penalties made against the contractor.
We also interviewed the Contracting Officer, the manager of FEMA's
Program Management Office, the COTRs for each contract we reviewed - as
well as contractors.
We conducted this performance audit in Atlanta, Georgia; Lanham,
Maryland; and Washington, D.C., from March 2007 to June 2008 in
accordance with generally accepted government auditing standards. Those
standards require that we plan and perform the audit to obtain
sufficient, appropriate evidence to provide a reasonable basis for our
findings and conclusions based on our audit objectives. We believe that
the evidence obtained provides a reasonable basis for our findings and
conclusions based on our audit objectives.
[End of section]
Appendix II: National Flood Insurance Program Statistics, 1997-2006:
Under the NFIP, homeowners with mortgages from federally regulated
lenders on property in communities identified to be in high-risk flood
zones are required to purchase flood insurance for at least the
outstanding mortgage amount, up to the maximum policy limit of
$250,000. NFIP covers both residential and commercial properties.
However, residential (defined as single-and two-four family dwellings)
and condominium properties account for more than 90 percent of all
policies in force. Optional, lower-cost coverage also is available
under NFIP to protect homes in zones designated as moderate-to low-
risk. In addition to building coverage, NFIP offers policies to protect
personal property ("contents") against flood damage, up to a maximum of
$100,000.
The tables and figures in this appendix provide selected information
for the period we analyzed on:
* policies in force;
* insurance coverage amounts;
* premiums;
* claims, and;
* losses by calendar year.
Tables and figures (see tables 9, 10, 11, 12, 13, 14, 15, and 16 and
figures 7 and 8) in appendix II include all data unless specified
otherwise. In addition, we provide data on estimates for historical
loss years. The estimates are used as one indicator to help set annual
premiums in NFIP that would be sufficient to pay for the average level
of losses that occurred in past years.
Table 9: Number and Percentage Change for Policies in Force, 1997-2006:
Calendar year: 1997;
Number of policies in force: 3,962,077;
Percentage change: Baseline.
Calendar year: 1998;
Number of policies in force: 4,114,319;
Percentage change: 3.84%.
Calendar year: 1999;
Number of policies in force: 4,206,914;
Percentage change: 2.25.
Calendar year: 2000;
Number of policies in force: 4,255,425;
Percentage change: 1.15.
Calendar year: 2001;
Number of policies in force: 4,360,678;
Percentage change: 2.47.
Calendar year: 2002;
Number of policies in force: 4,406,664;
Percentage change: 1.05.
Calendar year: 2003;
Number of policies in force: 4,447,774;
Percentage change: 0.93.
Calendar year: 2004;
Number of policies in force: 4,558,696;
Percentage change: 2.49.
Calendar year: 2005;
Number of policies in force: 4,827,181;
Percentage change: 5.89.
Calendar year: 2006;
Number of policies in force: 5,404,952;
Percentage change: 11.97%.
Source: GAO analysis of NFIP data.
[End of table]
Table 10: Maximum Level of Flood Insurance Available, Regular Program,
1997-2006:
Building type: Noncondominium and nonunit owner: Single-