Post-Katrina Insurance Issues Surrounding Water Damage Exclusions in Homeowners' Insurance Policies

Prepared for Members and Committees of Congress

In the aftermath of Hurricanes Katrina and Rita, homeowners in Louisiana, Mississippi, and
Alabama have protested what they view as inappropriate obstacles to the payment of their
property damage insurance claims. When insurance adjustors and damage experts assessed the
properties damaged by the 2005 storms, they were faced with the issue of allocating damages
between wind (a covered loss) and flood (an excluded loss). The delays and economic uncertainty
that this activity has engendered have raised financial and legal issues for insurers, as well as
homeowners and businesses along the Gulf Coast region.
The solution for some of the insureds has been to litigate with their insurers under a variety of
legal theories with the objective of avoiding the standard “water damage” exclusion—a move that
would arguably allow insureds to receive coverage for water damage under their homeowners’
policies. Many of these theories rely on the principle that any ambiguity in the insurance policy
should be construed in favor of the policyholder. In addition, even where the water damage
exclusion is enforced, difficult factual and legal disputes have arisen relating to the allocation of
damage between covered wind damage and excluded flood damage.
In the aftermath of the devastating 2005 hurricane season, three broad policy issues for the 110th
Congress have emerged related to post-Katrina economic uncertainties: (1) the massive insured
and uninsured property losses and their impact on Gulf Coast property insurance markets and
rebuilding after Katrina, (2) assertions that insurers have shifted the cost of damages onto the
federal flood program and U.S. taxpayers, and (3) unreliable government flood maps that are used
in decision making by homeowners for purchasing insurance.
Post-Katrina insurance claims litigation and the economic uncertainty it generates for consumers
and insurers raise concerns about the volatility in the legal environment in terms of post-event
judicial interpretations in the scope of insurance coverage. What should be done to mitigate the
economic consequences of future floods? Are the American people and policymakers ready to
address perceived weakness in the U.S. floodplain management policy? Should the nation forgo
development of its floodplains (a new policy initiative) or continue along the current path
embodied in the NFIP?
Finally, insurance analysts have observed that a large percentage of those eligible to buy federally
subsidized flood insurance do not. What could or should be done about this? These and other th
policy questions have been examined in the 110 Congress. On February 28, 2007, the House
Committee on Financial Services, Subcommittee on Oversight and Investigation, held a hearing
on insurance claims payment processes in the Gulf Coast after the 2005 Hurricanes. On
September 27, 2007, legislation—H.R. 3121, the Flood Insurance Reform and Modernization Act
of 2007—was passed in the House. A similar bill (S. 2310) desinged to strengthen the solvency of
the NFIP is pending in the Senate.. This report will be updated as events warrant.

Introduc tion ..................................................................................................................................... 1
The Issues........................................................................................................................................3
Overvi ew ....................................................................................................................... ............ 3
Impact of Hurricane Katrina on Insurance Markets..................................................................5
Post-Katrina Economic Uncertainty for Policyholders and Insurers......................................10
Issues of Contention..........................................................................................................10
Insurance Policy Language and Coverage....................................................................................12
Federal Flood Insurance..........................................................................................................12
Standard Flood Insurance Policy (SFIP)...........................................................................13
Standard Homeowners’ Insurance...........................................................................................14
Insuring Clauses................................................................................................................15
Named-Perils v. Open-Perils Policies...............................................................................15
Water Damage Exclusion..................................................................................................16
Claims Adjustment, Causation, and Policy Language Interpretation............................................16
Causation Battles Over Hurricane Katrina Claims..................................................................18
Concurrent Causation........................................................................................................18
Proximate Cause...............................................................................................................19
Efficient Proximate Cause................................................................................................19
Anti-Concurrent Causation Clauses..................................................................................19
Unfair Claims Practices State Laws........................................................................................20
Summary of Katrina Litigation.....................................................................................................20
“Valued Policy” Statutes and Related Litigation.....................................................................21
Other Litigation.......................................................................................................................22
Criminal Investigation of Insurer’s Claim Handling Practices...............................................23
Legislative Response.....................................................................................................................23
Concluding Observations..............................................................................................................25
Table 1. Ten Most Costly Catastrophes in the United States...........................................................6
Table 2. National Flood Insurance Program Data for Hurricanes Katrina and Rita........................6
Author Contact Information..........................................................................................................26

Hurricane Katrina, the storm surges it produced, and the subsequent levee failures caused
devastating flooding in communities located along the Gulf of Mexico Coast. Katrina’s winds
destroyed or substantially damaged many of the properties located on the Mississippi Gulf Coast,
as well as property located further inland from the immediate coastal areas. Tens of thousands of
homes and businesses in Louisiana, Mississippi, and Alabama were damaged as a direct result of
Katrina’s storm surge. Many thousands more homes were condemned and left empty due to
exposure to days and even weeks of soaking in often-contaminated flood waters.
Modern homeowners’ and business owners’ policies are often issued on an “open perils” or
“special causes of loss” basis which provides that all direct physical loss or damage to insured
property is covered except as specifically excluded. Some policies are still issued on a “named
perils” or “specified perils” basis by which the insurer promises to insure covered property only if
damaged by listed perils and subject to certain exclusions. Under either approach, damage by
wind is typically covered while water damage from flooding, wind driven water, storm surge,
seepage or through openings in the building (not cause by other damage) is typically specifically
There are three main reasons that insurers utilize exclusions: (1) the excluded peril or property is
more appropriately insured under a different insurance product or through an optional coverage
(e.g., exclusion in property policies for earthquake and mobile equipment); (2) insurance is not an
appropriate vehicle for transfer of a particular risk (e.g., exclusions in property policies for failure
of appropriate maintenance); or (3) the peril or property presents an unacceptable hazard to the
insurer (e.g., exclusions in property policies for nuclear or contamination event). The water
damage exclusion is thought to be the result of the first and third reasons. For many homeowners
and small businesses, the only explicit insurance coverage against flood damage is underwritten
by the National Flood Insurance Program (NFIP), but homeowners have not participated in 1
sufficient numbers for various reasons, as required by law.
Because of water damage exclusions and underutilization of available NFIP coverage, a
significant number of the thousands of properties damaged or lost through Hurricanes Katrina and
Rita were uninsured. These storms have spawned numerous class actions and other litigation
concerning insurance coverage for losses. Both uninsured and underinsured property owners have
focused on the water damage exclusions in homeowners’ policies, maintaining that they
apparently should be made whole through government flood insurance, homeowners’ insurance,
some combination of both, or the government, depending on each person’s particular insurance
For example, the Insurance Services Office (ISO) commercial policy wording (CP 10-30-04 02)
with respect to the flood exclusion provides the following:
We will not pay for loss or damage caused directly or indirectly by any of the following.
Such loss or damage is excluded regardless of any other cause or event that contributes
concurrently or in any sequence to the loss....

1 See Lloyd Dixon, Noreen Clancy, Seth A. Seabury,The National Flood Insurance Programs Market Penetration
Rate: Estimates and Policy Implications,” Rand Corporation, March13, 2006, located at

(1) Flood, surface water, waves, tides, tidal waves, overflow of any body of water, or
their spray, all whether driven by wind or not:
(2) Mudslide or mudflow;
(3) Water that backs up or overflows from a sewer, drain or sump; or
(4) Water under the ground surface pressing on, or flowing or seeping through;
( a) Foundations, walls, floors or paved surfaces;
( b) Basements, whether paved or not; or
( c) Doors, windows or other openings.
But if water, as described in g(1) through g(4) above, results in fire, explosions or sprinkler
leakage, we will pay for the loss or damage caused by that fire, explosion or sprinkler
Insurers take the position that these policies do not provide coverage for water damage resulting
from such events as levee breaches or storm surge because of the policy exclusion. For example,
the standard homeowners’ insurance policy language does not specifically identify “storm surge,”
but insurers insist that the breadth of the water damage exclusion encompasses such damage.
Insurance coverage disputes and litigation over the policy exclusions have ensued between
policyholders and their insurers. The key coverage issue has been and will be whether damage or
destruction to properties along the coastal regions was caused by winds (typically covered under
homeowners’ policies) or by the flooding accompanying the storm (typically excluded under
homeowners’ policies). With Katrina, the possibility exists that either wind or flood caused
damage—either separately or working in combination. Disputes over causation are inevitable.
Hundreds of millions and perhaps billions of dollars in potential insurance coverage will rest on
the outcome of decisions rendered in the courts.
On February 14, 2007, Mississippi’s largest provider of homeowners’ insurance, State Farm Fire
and Casualty Company, announced plans to suspend sales of new policies in the state because of 2
what the insurer claims is an increasingly unpredictable business and legal environment. The
decision is not expected to affect existing policyholders. A State Farm representative stated that
the decision to curtail writing new policies was a business decision to protect corporate assets in 3
response to an adverse legal and political environment in the state.
State Attorney General Jim Hood responded to State Farm’s announced plans by unveiling a
legislative proposal to the state’s Governor, insurance commissioner and the public that would
compel insurers to continue writing policies in the state. The proposal, which is based on similar

2 Steve Tucker, “State Farm Halts New Mississippi Business,National Underwriter Online News Service, February
14, 2007.
3 Mark E. Ruquet,Hood Calls State Farm on Bullying Tactics,” National Underwriter Online News Service, February
16, 2007.

legislation enacted in Florida, would require insurers writing auto insurance in the state to provide 4
homeowners’ and commercial insurance if they sell those policies elsewhere in the United States.
Following the State Farm announcement, the concern raised by insurance regulators and
policymakers is what to do about other insurers who might decide to retreat further from the Gulf
Coast region because of the uncertainty of pending legal battles. Consumer advocacy groups
expect other insurers to follow State Farm’s lead as a way to apply political pressure on the 5
Mississippi courts and legislature to deal with the insurance coverage issue.
This report provides an analysis of post-Katrina insurance issues in Louisiana, Mississippi, and
Alabama, particularly as they have related to (1) the magnitude and impact of flooding, (2) the
way private insurers settle claims, (3) the types and scope of policies and coverage, and (4) how
courts have determined whether coverage exists under property policies when both covered and
uncovered risks combine to cause a loss. The report does not provide legal advice, nor is it a
definitive assessment of the applicable law in each jurisdiction.

Private insurers play a central role in the functioning of the U.S. economy not only because of
their risk-transfer function, but also because of their ability to provide relatively easy access to
sources of capital. Instability in the availability and price of coverage, interruptions in the
payment of claims immediately after a disaster, and the large numbers of uninsured or
underinsured Americans have led to pressure for government intervention in property insurance
markets. In the wake of Katrina’s massive flooding and uninsured property losses, specific
concerns have been expressed in Congress not only about making sure that every insurance claim
owed is paid, but also that there is a Gulf Coast insurance market after all the litigation is
In 1983, as a result of adverse court decisions in which insurers were forced to pay flood-related
claims that insurers did not believe they were responsible to pay, the property and casualty
insurance industry revised its policy language in the exclusions in homeowners’ policies. Today,
almost all homeowners’ policies have a water damage exclusion with anti-concurrent causation 6
language which, according to insurers, should have eliminated coverage for flooding including
from storm surge and levee breaches. Not all water is excluded, just the water damage described
in the exclusion. There are still several instances in which water damage is covered, such as when

4 In addition to signing the new catastrophic insurance legislation, the Florida Governor signed an emergency order
suspending the right of insurers to cancel and non-renew pending full implementation of the new law. The Florida
Insurance Council, an insurance company trade group, initially challenged the validity of the Governors order, but
later withdrew the legal challenge. Events in Florida are not discussed in this report.
5 Joseph B. Treaster,State Farm Ends New Property Coverage in Mississippi,” New York Times, February 15, 2007, p.
6 The termanti-concurrent causation language refers to language that is typically inserted in an insurance policy to
make it as clear and unambiguous as possible that a specific risk is not covered. Insurers use this type of language in
contracts in an attempt to prevent policyholders from seeking coverage for losses they never intend to cover.

the pipes in a home freeze and burst, when a roof is ripped off the home and rain water comes in,
or the fire department hoses down the house.
Particular policy phrases and their interpretation frequently are at the center of insurance-based
litigation. The central question is how the courts will interpret the water damage exclusion or
whether they will rule in favor of the claimants, forcing insurers to pay billions of dollars to repair
flood damage. Certain other insurance-related issues have arisen, such as concerns that some
insurers inappropriately billed the federal flood insurance program for claims that should have 7
been paid by the insurers’ wind policies.
Some Members of the 110th Congress support efforts to investigate the Katrina claims practices of 8
insurance companies that contract with the National Flood Insurance Program. On February 28,

2007, the House Committee on Financial Services, Subcommittee on Oversight and Investigation,

held a hearing on insurance claims payment processes in the Gulf Coast after the 2005
Hurricanes. In response to the hearing, the Government Accountability Office (GAO) recently
released a report that found that insurance coverage gaps and claims uncertainties can arise when 9
coverage for hurricane damage is divided among multiple insurance policies. GAO concluded
that greater transparency and oversight of wind and flood damage determinations are needed.
On September 27, 2007, legislation (H.R. 3121, the Flood Insurance Reform and Modernization
Act of 2007) was passed in the House to restore the financial solvency of the National Flood
Insurance Program. H.R. 3121 includes a provision to add optional multiple peril coverage for flood
and windstorm losses. A similar bill (S. 2310) is pending in the Senate, but it does not include the
provision for the optional windstorm coverage.
Other Members of Congress support legislation that would create a federal catastrophe fund
backed by state funds to shore up the private insurance industry in the event of a mega-
catastrophe. Still other Members support the repeal of the insurance industry’s limited federal
antitrust exemption to address what they insist is an anti-competitive industry. The exemption
from federal antitrust laws has allowed collaboration in the industry, such as development of 10
standardized policy forms.
The Louisiana, Mississippi, and Alabama Departments of Insurance, like those in other states,
have a right of review and approve policy forms before use. As with all other states, Louisiana
and Mississippi have for many years reviewed and approved flood and related water exclusions.
Departments of Insurance have also undertaken efforts to advise residents that most personal and
commercial property insurance policies do not cover flood and that a national program (NFIP) is
available to meet flood insurance needs. This frequent communication effort includes fact sheets

7 Matt Brady, “U.S. Representative Taylor Seeks Probe of Insurers, National Underwriter Online News Service,
January 12, 2007.
8 Arthur D. Postal, “Gulf Congressmen Attack Insurers on Many Fronts, National Underwriter Online News Service,
January 19, 2007.
9 U.S. Government Accountability Office, National Flood Insurance Program: Greater Transparency and Oversight of
Wind and Flood Damage Determinations are Needed,, visited on January 30,
10 For a brief discussion of what the McCarran-Ferguson Act does and does not cover, as well as some McCarran-
related legislation in the 109th Congress, see CRS Report RL33683, Courts Narrow McCarran-Ferguson Antitrust
Exemption for “Business of Insurance: Viability ofState Action Doctrine as an Alternative, by Janice E. Rubin.

and press releases (Louisiana—2002, 2003, 2004, 2005; Mississippi—2000, 2002; Alabama—

2002, 2004, 2005).

In addition, the federal government through the NFIP has for years undertaken a multi-media
campaign to advise homeowners and business owners that most insurance policies do not cover
flood—but that federal coverage is available. For homeowners’ insurance, federal flood insurance
is typically less than $1,000 per year (although pricing varies depending on the value insured and
location of the property). For businesses, the federal flood insurance premium is typically a
couple of thousand dollars, again depending on the amount of coverage and location of the
Litigation between policyholders and their insurers over the interpretation of “water damage”
exclusion and the “anti-concurrent causation” clauses could continue for months and even years.
The outcome of legal disputes will likely determine how losses are eventually apportioned among 11
the National Flood Insurance Program (NFIP), private insurers, individuals, and businesses.
In the aftermath of Hurricanes Katrina and Rita, three broad issues have emerged related to post-
Katrina economic uncertainties:
• the massive insured and uninsured property losses and their impact on Gulf Coast
property insurance markets and rebuilding after Katrina;
• assertions that insurers have shifted the cost for damage to the federal flood
program and FEMA for U.S. taxpayers to pay the bill—i.e., potentially huge
uninsured property losses and the denial of thousands of Katrina wind claims
where insurers invoked the “water damage exclusion” and “anti-concurrent
causation” clause in homeowners’ policies;
• unreliable government flood maps that are used by homeowners, lenders, and
realtors to determine whether flood insurance was needed.
First, Hurricanes Katrina, Rita and Wilma resulted in massive property losses for both private
insurers and the National Flood Insurance Program (NFIP). In addition, the total value of
uninsured property damage and business interruption caused by Katrina has been estimated at 12
$135 billion. Congress has already appropriated $109 billion for disaster relief and recovery aid 13
to affected communities, plus more than $8 billion in tax relief.
Table 1 shows that private insurers paid $60.5 billion in insured property damages caused by
Hurricane Katrina, Rita, and Wilma. This amount does not include claims filed under the NFIP.
Insurers are still assessing losses in terms of how they will affect ultimate claim payments for
losses in 2005, including the total cost of litigating and settling “wind v. flood” cases.

11 Bill Mellander, “Payouts Hinge on the Cause of Damage,” New York Times, August 31, 2005, p. C5.
12 Council of Economic Advisers, Economic Report of the President, February 2007, p. 120.
13 Matt Fellowes and Amy Liu, “Federal Allocations in Response to Katrina, Rita and Wilma: An Update,” The
Brookings Institutions, located at
20060712_katrinafactsh eet.pdf#search =%22communit y%20developmen t%20block%20grant%20and %20katrina%20fl

Table 1. Ten Most Costly Catastrophes in the United States
Insured Loss ($ millions)
Rank Date Peril Dollars When Occurred In 2007 dollars
1 Aug. 2005 Hurricane Katrina $41,100 $43,625
2 Aug. 1992 Hurricane Andrew 15,500 22,902
3 Sept. 2001 World Trade Center, Pentagon Terrorist Attacks 18,800 22,006
4 Jan. 1994 Northridge, CA earthquake 12,500 17,485
5 Oct. 2005 Hurricane Wilma 10,300 10,933
6 Aug. 2004 Hurricane Charley 7,475 8,203
7 Sept. 2004 Hurricane Ivan 7,110 7,803
8 Sept. 1989 Hurricane Hugo 4,195 7,013
9 Sept. 2005 Hurricane Rita 5,627 5,973
10 Sept. 2004 Hurricane Frances 4,595 5,043
Source: Insurance Services Office (ISO); Insurance Information Institute.
Table 2 shows Hurricanes Katrina and Rita caused a record 226,419 flood insurance claims of
which 164,615 individual claims were paid. Some 62,849 damaged homes were not covered by
flood insurance and 58,413 claims were closed without payment. According to the Federal
Emergency Management Agency (FEMA), as of December 31, 2007, the NFIP paid $19.28
billion in claims payments—an amount that far exceeds the aggregate amount of claims paid in
the history of the program.
Table 2. National Flood Insurance Program Data for Hurricanes Katrina and Rita
(As of January 5, 2007)
State and Number of Flood Number of Homes Number of Household Number of Flood Insurance Claims
Disaster Insurance Claims Damaged Without aIndividual Claims Closed Without
Number Received Flood Insurance Fully Paid Payment
Florida, DR-9,021 733 5,814 3,070
Louisiana, DR - 177,827 32,768 2 127,330 47,942
Mississippi, DR 19,006 26,716 17,159 1,519
Alabama, DR - 5,725 1,743 5,123 557
Texas, DR-3,930 889 1,721 2,111
Louisiana, DR-10,910 N/Ab 7,468 3,214
Total 226,419 62,849 164,615 58,413
Source: FEMA Office of Federal Affairs

a. Data are as of March 28, 2006;
b. Louisiana disaster numbers DR-1603 and 1607 combined.
Unprecedented losses in the NFIP have led to unprecedented borrowing from the U.S. Treasury.
As of December 31, 2007, the NFIP had borrowed $17.53 billion from the U.S. Treasury, and
must be repaid with interest. The Congressional Budget Office calculates that FEMA is unlikely 14
to repay the funds borrowed to pay 2005 hurricane-related claims withing the next 10 years. th
Consequently, the 110 Congress has been called upon to overhaul the National Flood Insurance
Program in order to address: (1) program’s financial solvency; (2) indebtedness to the U.S.
Treasury; (3) magnitude of uninsured flood damage due to the widespread noncompliance with
the mandatory flood insurance purchase requirements; and (4) controversy surrounding the
practice of contracting with private insurers for policy adjusting and servicing.
Legislative efforts are now underway in Congress to reform the NFIP. On September 27, 2007,
the full House approved H.R. 3121 to restore the financial solvency of the national flood
insurance program. Significantly, H.R. 3121 would allow for the purchase of optional insurance
that would cover flood and windstorm losses and extend the NFIP five years through September
30, 2013. On November 1, 2007, Senator Dodd introduced S. 2284, a flood insurance reform bill
similar to H.R. 3121 in terms of increasing the amount of premiums collected and reducing the
cost of expected claims. S. 2284, however, would forgive the program’s outstanding debt to the
Second, Hurricane Katrina damaged or destroyed thousands of homes and businesses that were
covered for wind damages, but not water damage. Insurers have reportedly denied thousands of
Katrina wind claims by assigning all Katrina damages to flooding covered by the NFIP and not to
their own windstorm policies. Insurers say they are simply invoking the “water damage”
exclusion and “anti-concurrent causation” clause in homeowners’ policies. Policyholder
advocates claim insurers have a conflict of interest because they are able to shift the cost for
damage to the federal flood program rather than to themselves. Further, insureds doubt whether 15
FEMA provides proper oversight of NFIP’s Write Your Own (WYO) insurers to ensure they are
fulfilling their contractual obligations to fairly adjust flood claims, particularly those involving
combined wind and water damage. Finally, critics charge that, while the states have a role in
regulating the claims adjusting process, the “Unfair Claims Practices” statutes are not being
adequately enforced by the states.
On the other side of this issue, many homeowners who were eligible for low cost NFIP flood
coverage (including those required by law to purchase it) declined this available coverage despite
state and federal educational efforts. Many insurers—and several recent editorials—assert that
policyholders’ claims of flood coverage under homeowners’ policies with flood exclusions are
post-event rationalizations for poor personal planning decisions.

14 See Letter from Donald B. Marron, Acting Director of Congressional Budget Office, to Honorable Judd Gregg,
Chairman, Committee on the Budget, May 31, 2006, located at
15 WYO insurers are private insurers who have agreed to sell and service NFIP policies, and adjust flood insurance
claims under a contractual agreement with FEMA.

Researchers have suggested that millions of families are now living in flood-prone areas without
adequate insurance protection. This is despite that fact that structures in areas with at least a 1%
chance of flooding in any given year—the so-called “100-year” flood or Special Flood Hazard
Areas (SFHAs)—are required to have flood insurance if they have a loan from a federally insured
or regulated lender. The insurance must be in an amount of at least equal to the outstanding 16
principal balance of the loan or the maximum available under the NFIP, whichever is less. This 17
requirement is based on two federal laws: the Flood Disaster Protection Act of 1973 and the 18
National Flood Insurance Reform Act of 1994.
According to a Rand Corporation nationwide study, about 49% of single-family homes in SFHA
are covered by flood insurance (the market penetration rate), with substantial variations across 19
geographic regions. In addition, about 1% of homeowners in non-SFHAs purchase flood 20
insurance coverage. The purchase of flood insurance is voluntary outside SFHAs.
Researchers have explored factors influencing insurance purchase decisions.21 They include the
• Millions of homeowners incorrectly believe that their standard homeowners’
policies automatically provide coverage against flooding, when in fact an
additional flood policy is needed.
• People have misperception of risk and tend to underestimate their chances of
being disaster victims and do not purchase flood insurance. They have difficulty
dealing with probabilistic information for small likelihood events because they 22
need a context in which to evaluate the data. Homeowners might think it is not
a good investment when comparing the insurance price of coverage per dollar
against their estimate of the probability of total loss, which they assume
approaches zero.
• There are economic disincentives to buying flood insurance because of “free”
federal and charitable disaster assistance and the tax-deductibility of flood 23
losses. Property owners often believe that disaster relief (i.e., Small Business

16 These property owners or loans include (1) loans from federally regulated lending institutions; (2) loans that are
purchased by the Federal National Mortgage Association (Fannie Mae) or the Federal Home Loan Mortgage
Corporation (Freddie Mac); and (3) property owners who receive federal financial assistance for acquisition or
construction purposes in SFHAs in communities that participate in the NFIP.
17 P.L. 93-234; 87 Stat. 975.
18 P.L. 103-325; 108 Stat. 2255.
19 See Lloyd Dixon, Noreen Clancy, Seth A. Seabury,The National Flood Insurance Programs Market Penetration
Rate: Estimates and Policy Implications,” Rand Corporation, March 13, 2006, located at
20 Ibid.
21 Howard Kunreuther, “Has the Time Come for Comprehensive Natural Disaster Insurance?” in On Risk and Disaster:
Lessons from Hurricane Katrina, Roland J. Daniels, Donald F. Kettle, eds. (Philadelphia: University of Pennsylvania
Press, 2006), p. 175.
22 Howard Kunreuther and M. Pauly, “Rules Rather than Discretion: Lessons from Hurricane Katrina, Journal of Risk
and Uncertainty, 2006, vol. 33, pp. 101-116.
23 Howard Kunreuther and M. Pauly, “Neglecting Disaster: Why Dont People Insure Against Large Losses?” Journal
of Risk and Uncertainty, 2004, vol. 28, pp. 5-21.

Administration (SBA) low interest loans and grants) will compensate them for
flood losses.
• The NFIP can undercompensate for losses given the high property (and land)
values for coastal properties and the NFIP’s maximum coverage of $250,000 for
structure and $100,000 for contents.
• Property owners sometimes self-protect or self-insure—a substitute for formal
insurance—by, for example, elevating their property above the base flood
elevation (BFE), which serves to reduce both the probability of loss and the size
of the potential costs.
• Homeowners might simply cancel their flood coverage without lenders taking
• Homeowners with mortgages issued by non-federally regulated lenders are not
subject to the mandatory flood purchase requirements.
In response to low levels of compliance with the mandatory purchase requirements and the
financial challenges facing federal insurance program, some insurance market analysts have 24
proposed an overhaul of the NFIP to address these and other programmatic issues. Although
FEMA does have an interest in ensuring high levels of compliance, the agency must rely on
federally regulated lenders, government sponsored enterprises (GSEs), and federal agencies that
provide financial assistance for construction and acquisition of property in SFHAs to implement
and monitor compliance with the mandatory purchase requirement. Some propose requiring all
U.S. homeowners to purchase flood insurance, possibly the way liability insurance is required for
automobile owners.
Third, thousands of residents along the Gulf Coast who relied on government flood maps to
determine who must buy flood insurance were arguably unaware of their actual flood risk and th
uninsured when Hurricanes Katrina, Rita, and Wilma struck. An issue for the 110 Congress,
therefore, might be to decide what to do about flood maps that may be incorrectly measuring risk
and, therefore, not capturing the nation’s exposure to flood risk. Critics of FEMA’s flood map
modernization efforts say these activities have been underfunded and lack focus.
Many people in Hurricane Katrina’s path did not have flood insurance because, according to 25
government flood maps, they were not in a floodplain and did not need coverage. On October

18, 2005, the Department of Homeland Security’s Inspector General released a report that raised 26

concerns about the accuracy of flood maps. The report concluded that outdated flood maps
placed homeowners and residents at physical and financial risk because many people living in
high-risk areas did not know of the dangers and, therefore, might not have chosen to participate in
the NFIP.
Again, according to Rand Corporation, about one-third of all flood claims occur outside of
SFHAs, yet only 1% of property owners residing in these areas purchase flood insurance from the

24 For more information on flood insurance reform see CRS Report RL34052, The Flood Insurance Reform and
Modernization Act of 2007: A Summary of Key Provisions, by Rawle O. King.
25 Office of Inspector General, Department of Homeland Security, “Challenges in FEMAs Flood Map Modernization
Program, Report No. OIG-05-44, September 2005, located at
26 Ibid.

NFIP. Communities in harm’s way might not be adequately enforcing strong building codes or
land use zoning ordinances, which would have reduced the amount of wind-related damage.
In the case of New Orleans—a city that experienced massive flooding and an historic 29-foot
storm surge—FEMA flood maps incorrectly made the assumption that levees and flood walls 27
would withstand the storm surge and protect the residents from flooding. Many property owners
in Katrina-impacted areas outside government-designated flood zones whose homes were
destroyed were not subject to the mandatory purchase requirement and, therefore, did not have
adequate coverage. Efforts are underway at FEMA under its flood map modernization program to
develop accurate digital flood plain maps that are consistent in terms of data collection standards
and analytical methods.
In the aftermath of Hurricanes Katrina and Rita, thousands of homeowners in Louisiana,
Mississippi, and Alabama face significant obstacles in getting their property damage insurance
claims paid. When insurance adjustors and damage experts assessed properties damaged by the
2005 hurricanes, they were faced with the issue of allocating damages between wind (a covered
loss) and flood ( an excluded loss). The ambiguity and economic uncertainty that this activity has
engendered has raised financial and legal issues for insurers, as well as homeowners and
businesses along the Gulf Coast region. The solution for some of the insureds has been to litigate
with their insurers, seeking to declare the “water damage exclusion” clause in their policies
unenforceable—a move that would arguably then allow insureds to receive coverage for flood
damage under their homeowners’ policies. Allocating damages between covered wind damages
and excluded flood damages has become a question of fact to be decided by the courts.
Insurers attempt to separate the wind damage (covered) from the flood damage (excluded)
through physical examination of the location. In many areas of New Orleans, for example, the
house contains a clear water-line with no or minimal roof damage. Along the Mississippi coast,
many homes are completely destroyed with only a slab remaining. In those cases, the factual
investigation turns on an analysis of weather conditions and the construction of the home to
determine the extent of wind damage that occurred prior to the destruction of the home by
excluded storm surge. Policyholders and insurers can have different views on these facts which
result in litigation.
From the industry’s perspective, claims are settled in the same manner they always have been
with regard to the water damage exclusion that they assert has been well known among
homeowners and state regulators for years. Insurers have argued that they settle claims fairly and
in accordance with policy language and the historical treatment in court cases of storm surge
associated with flooding during a hurricane. This is why, insurers insist, the federal government
offers flood insurance. In addition, insurers note that they did not cover, and therefore price, the
flood risk in the homeowners’ policies, and hence did not set aside appropriate loss reserves to 28
pay such claims. Further, insurers maintain that the uncertainty associated with the court’s ex

27 Peter Whorisky, “Risk Estimates Led to Fewer Flood Policies,” Washington Post, October 17, 2005, p. A 1.
28 Ibid.

post re-interpretation of insurance policy terms and language substantially increases their risk—
i.e., paying flood insurance claims after the loss without actuarially accounting for this event
before the loss could jeopardize their financial solvency. If they raise rates to reflect the increased
risk, their customers will pay more for coverage that is now provided by the NFIP.
Policyholders and their advocates have charged that insurers rely upon the “flood exclusion” and
anti-concurrent causation policy language to not pay claims on homes damaged by a combination
of wind and flood. From their perspectives, insistence on the unenforceability of the “flood
exclusion” and the anti-concurrent clause is consistent with “reasonable expectation” since they
thought they had “full protection.”
Post-Katrina insurance claims litigation and the economic uncertainty it generates for consumers th
and insurers have raised at least four interrelated policy issues for the 110 Congress: insurance
coverage disputes; disaster and the law; flood plain policy; and insurance reforms.
There may be congressional oversight and investigations into the handling of Katrina-related
insurance claims, and possible legislative efforts to modify the industry’s antitrust exemption 29
under the McCarran-Ferguson Act of 1945 and state supervision and regulation of the business th
of insurance. Members of the 110 Congress have the option to address the volatility in the legal
environment in terms of post-event judicial interpretations in the scope of insurance coverage and
legislative or regulatory “lock-ins” of capacity in the months and even years after a storm. After
Katrina and Rita, for example, in states like Florida and Louisiana, the insurance commissioners
declared a state of emergency in the state’s insurance market and suspended certain statutes and
regulations regarding policy cancellations, non-renewals, reinstatements, and claim filings. These
emergency declarations have had the effect of “locking in” the investor capital that stands behind
the policies sold in the state. Some insurance experts have suggested that legal and regulatory
volatility tends to discourage rather than encourage the retention and expansion of capital
commitments in the affected states.
Katrina demonstrates apparent gaps in the legal system and its ability to respond to events of this
magnitude. This issue is important if communities are to be able to rebuild and recover in a timely
manner after a major catastrophe. In the short term, Congress might consider oversight and
investigation hearings into post-Katrina insurance claims issues and the regulatory impact this
could have on the insurance market. In the long term, businesses, including the insurance
industry, rely on a predictable legal regime that will operate efficiently in an emergency situation.
In anticipation of another Katrina-sized natural disaster, Congress has the option to consider the
enactment of a nationwide body of “disaster law”so the nation might be better prepared to rebuild 30
and recover after a major catastrophe.

29 P.L. 79-15; 59 Stat. 33
30 See Daniel A. Farber, “Disasters and the Law: Katrina and Beyond” (Aspen Publishers: 2006.)

What should be done and not done to recover from flood damage and mitigate the consequences
of future floods? The aftermath of a disaster often presents the opportunity to address multiple
long-standing problems. After Hurricane Katrina, the floodplain policy response seems to have
been to build “bigger and better” flood protections systems—i.e., take a nationwide inventory of
the structural flood controls (levees), reinforce and strengthen them, and expand requirements for
insuring residual risks behind levees and dams. History (1920s-1960s and 1993 Midwest floods)
has showed that relying on structural means of controlling floods, such as building levees and
dams, is not always the best flood control policy. Two broad policy options have been suggested
that would (1) surrender land to the water, a new idea that would require forgoing development of
floodplains and property buyouts; or (2) continue along the current path embodied in the NFIP
(i.e., risk assessment or mapping, floodplain management, and insurance protection), but
strengthen zoning laws and construction standards, modernize flood maps, and enforce existing
mandatory flood insurance purchase requirements.
A large percentage of those eligible to buy federally subsidized flood insurance do not purchase
the coverage. The NFIP subsidizes insurance rates for about 26% of policies, generally high-risk
buildings built before NFIP floodplain regulations were established in their community. As a
long-standing public policy, the federal government forgoes significant premium income because
of policy subsidies. Moreover, losses associated with subsidized properties account for 25% to
30% of all claims losses. What is the solvency and regulatory impact of massive Katrina-related
flood losses on the NFIP? How effective has the program’s mandatory flood insurance purchase
requirement been in increasing market penetration and reducing future flood losses?

Private insurers generally do not cover the flood hazard because of the problem of adverse 31
selection, and the perceived unprofitability and volatility of this line of insurance due to the
absence of tools for technical risk rating or portfolio diversification. Since 1968, this gap in
coverage has been filled by the purchase of federally subsidized flood insurance. Many
homeowners incorrectly believe that their standard homeowners policies automatically provide
coverage against flooding, when in fact an additional flood policy will be needed.
This section provides an analysis and comparison of both federal flood insurance and the NFIP’s
standard homeowners’ insurance policy, and the nature and types of the insuring clauses,
including the “water damage” exclusion provision.
Federal flood insurance is available to residents only in communities that agreed to institute
floodplain management strategies designed to reduce future flood losses. The federal government

31 Adverse selection is the tendency of people who have a greater perceived probability of loss than does the average
person to seek insurance.

established certain minimum building and development standards for floodplain construction that
the communities have to adopt in order to participate in the NFIP. Under the National Flood 32
Insurance Act of 1968, the federal government is required to map the nation’s floodplain. In
order to do so, FEMA determines flood risk through various sources specific to each community. 33
This flood risk information is then delineated by zones on Flood Insurance Rate Maps (FIRMs).
The area of flood hazards on these maps is the Special Flood Hazard Area (SFHA), which is
defined as an area of land that experiences a 1% chance of being flooded in any given year (also
known as the base flood or 100-year flood).
The NFIP offers three Standard Flood Insurance Policy (SFIP) forms: Dwelling Policy; General
Property Policy; and Residential Condominium Building Association Policy. The Dwelling Policy
is used to insure residential structures and their contents and is issued to homeowners (including
those in condominium units, manufactured mobile/trailer homes, townhouse structures, and
timeshares), residential renters, or owners of residential buildings containing two to four units.
The Dwelling Form offers coverage for building property, up to $250,000, and personal property
(contents), up to $100,000. Contents coverage must be purchased separately.
The General Property Policy is issued to owners of residential buildings with five or more units,
owners or lessees of nonresidential structures, such as hotels, apartment buildings, schools,
commercial structures, cooperative associations, and their contents. Coverage is available up to
$500,000 for non-residential buildings and their contents.
The Residential Condominium Building Association Policy (RCBAP) is issued to residential
condominium building associations to cover the entire building under one policy. The policy
covers all units, improvements within the units, and personal property owned in common.
Eligible structures under the RCBAP include high-rise and low-rise condominium buildings and
condominium associations.
The SFIP is not a guaranteed replacement cost policy that pays the to rebuild regardless of the
limit of liability in the event of a total loss. Instead, the SFIP policies pay the replacement cost of
actual damages, up to the policy limit. In other words, flood insurance does not pay more than the
policy limit.
The NFIP defines a flood as
general and temporary condition of partial or complete inundation of two or more acres of
normally dry land area or of two or more properties from: overflow of inland or tidal waters,
unusual and rapid accumulation or runoff of surface waters from any source, mudflow, or
collapse or subsidence of land along the shore of a lake or similar body of water as a result of

32 P.L. 90-448, 82 Stat 573.
33 In order to assess a communitys flood risk which is delineated on FIRMs, FEMA uses historical flood data,
including the communitys rainfall and river-flow, topography, wind velocity, tidal surge, flood-control measures,
development, and other factors.

erosion or undermining caused by waves or currents of water exceeding anticipated cyclical
The SFIP covers physical damage to the building or personal property “directly” caused by a
flood. It does not cover: damage caused by moisture, mildew or mold; loss of currency and
valuable papers such as stock certificates; living expenses such as temporary housing; and,
financial losses caused by business interruption or loss of use of insured property.
The property insurance industry uses standardized policy language to provide uniformity in
coverage and consistency in legal outcomes. The basic forms are developed and written by the 34
Insurance Services Office Inc. (ISO). The standard homeowners’ insurance (HO) policy that
ISO publishes combines various personal insurance protections which can include losses on the
home, its contents, loss of its use (additional living expenses), as well as liability coverage in the
event the homeowner is sued and found legally responsible for damages.
There are seven types of ISO standardized HO forms in general and consistent use. Of these, HO-

3 is the most common policy, followed by HO-4 and HO-6. Others that are less used are HO-1,

HO-2, HO-5, and HO-8. The HO-3 is a special homeowners policy—called open-perils—that
covers all perils except those specifically excluded by the policy such as earthquakes, floods, Acts
of God, or war. Special insurance can be purchased for these possibilities, including flood
insurance and earthquake insurance. The HO policy might also contain options, called riders,
floaters, or endorsements, that can provide additional coverages for such items as art or coin
collections for an additional premium.
State insurance departments are responsible for reviewing and approving all insurance policy
forms and rates before the policy can be lawfully used by an insurer. The approval process
includes an examination of every word, sentence, and paragraph in the policy. Before the
regulators are given the policy form for approval, however, it is standard practice in the industry
for insurers to consult with actuaries and insurance underwriters to develop the policy language
so that pricing is commensurate with the related coverage. Economic efficiencies are realized in
this activity by utilizing certain industry organizations, like the ISO, to develop the specific
provision in standard-form insurance policies and to file with the each state insurance department
so that individual insurers can use them without expending money and time that would be
required to get the policy forms approved in every state in which they operate.
By standardizing the HO policy, insurers hope that court interpretations of standard coverage
forms provide consistent treatment of claimants. Thus, when a court determines the meaning of a
word, phrase, or clause in a standard coverage form, that interpretation triggers a rewrite or
adjustment to the policy language.

34 The Insurance Services Office, Inc. (ISO) is an insurance organization that provides statistical information, actuarial
analyses and consulting, policy language, and technical services to insurers.

Property and casualty (p/c) insurance policies are classified by causation, and contain in their
insuring clauses the words, “loss caused by...” or their equivalent. The insuring clause in the HO-
3 policy is a statement of the promises the insurer makes to the insured as far as what perils and
exposures are covered; it varies greatly from policy to policy. The problem is that when an
insured cause (wind) joins with one or more additional causes (flood), which may be uninsured or
may be insured under a separate contract, concurrent causation can be said to exist. It is this
concurrent causation that has generated so much litigation surrounding the “wind v. flood” legal
disputes. Litigation arises as to whether the damage was caused by an insured event or by an
event which is either excluded from coverage or not within the scope of the policy.
Property policies come in several types, including commercial and homeowners. These policies
are further divided by the type of risks insured: named-peril risk or open-peril risk. Named risk
policies insure against physical loss or damage caused by various risks specified in the policy
(e.g., fire, lightning, explosion, windstorm or hail, smoke, aircraft or vehicles, riot or civil
commotion, vandalism, sprinkler leakage, sinkhole collapse and volcanic action).
Open-peril policies, on the other hand, insure against physical loss or damage caused by any risk
that is not excluded or limited by the policy. The distinction between named-peril and open-peril
policies becomes especially important in litigation. Under the named-peril policies, courts have
held that a policyholder must first prove one or more of the named perils caused a loss. Insurers
must then prove that the loss is excluded in the policy. Under an open-peril policy, the
policyholder must prove damage occurred to the insured property. Insurers must establish that an
excluded risk caused the loss.
In some litigation addressing the water damage exclusion, plaintiffs have contended that the water
damage should be covered because it resulted from storm surge, which was not a specifically
excluded peril. A recent federal court decision held that “storm surge” was within the policy 35
definition of “flood” and the exclusion should apply. In another case involving water damage
but that specifically covered winds or hail, the court denied the insurer’s motion to dismiss, hold
that there was coverage for all damage caused by wind, and wind-driven rain, but that losses 36
directly resulting from storm surge were excluded by the water damage exclusion. Applying
Mississippi law in the Tuepker case, the judge also held that the exclusionary language in the
policy was invalid to the extent that it did not allow for consideration of the proximate cause 37
(covered wind or uncovered water) of the insured’s damage; and further that, at trial, if the
insured could prove its damage was sustained as a result of covered perils, its claim would be 38
Open-perils policies cover many perils not covered by named-perils policies; consequently, the
open-perils policy provides broader coverage than a named-perils policy, and carries a higher

35 Buente v. Allstate Prop. and Cas. Ins. Co., 2006 WL 980784 (S.D. Miss. 4/12/2006).
36 Tuepker v. State Farm Fire and Cas., 2006 WL 1442489 (S.D. Miss. 5/24/2006).
37 Ibid., p. 4.
38 Ibid., p. 6.

premium. Whether the policy is open-perils or named-perils, however, the coverage it provides
will have exclusions. Exclusions are an integral part of every insurance policy. The flood peril, in
particular, is excluded from HO policies because it is unusual and requires a separate rating.
In the event of hurricane-related flooding, the question of the cause of damage, whether wind or 39
water or wind and water is of considerable importance. The reason is that flood exclusion
language has been a standard feature of homeowners’ insurance policies since around 1968, when
the federal government’s National Flood Insurance Program (NFIP) was established.
Homeowners’ policies generally exclude coverage for flood damage. Insurers have not priced
and, therefore, not collected premiums to pay for flood coverage.
An insurance policy usually specifies what causes—called “hazards” or “perils”—are covered. It
also specifies what effects—called “losses” or “damages”—are covered and which perils and
losses are excluded from coverage. The insurance industry defines an “exclusion” as a provision
in an insurance policy that eliminates coverage for certain risks, people, property, classes, or
locations. The insurance policy language that specifically excludes flooding is found in the
“Perils Insured Against” and “Exclusions” section of the standard homeowners’ policy.

Two broad sets of post-Katrina claims-adjustment issues may be relevant to the 110th Congress.
First is the alleged adverse impact on insureds of computerized claims settlement systems and
products. Public interest advocacy groups have alleged that the insurance industry uses computer
programs, such as “Colossus” sold by Computer Sciences Corporation (CSC) or “Claims
Outcome Advisor” sold by the Insurance Services Office (ISO), to systematically underpay 40
homeowners claims. Their argument is that these systems allow insurers to calibrate the amount
of savings they want to generate to the detriment of their insureds.
Claims adjustment under the NFIP is different from that in the private homeowners’ insurance
market. The adjustment of federal flood insurance claims is conducted by property and casualty
insurers who write and service NFIP policies and claims under a contractual agreement with the
NFIP’s Write Your Own (WYO) Program. Over 95% of NFIP’s policies are written under this
program. Private insurers enter into a “Financial Assistance Subsidy Arrangement” whereby they
agree to issue flood policies in their own names and take responsibility for policy administration,
claims processing, marketing, and sales. Private insurers handle all claims issued in their names,
and adjust and settle flood loss claims consistent with their general claims practices. In adjusting

39 The controversial issue involving the allocation of damage between wind and water is not present in the commercial
insurance marketplace because flood/water damage is covered. To the extent an issue might arise, it is with respect to
sublimits for damage from flood. For example, a commercial building could be insured for $200 million but with a $30
million sublimit for flood.
40 See J. Robert Hunter, Property/Casualty Insurance in 2007: Overpriced Insurance, Underpaid Claims, Declining
Losses and Unjustified Profits, January 8, 2007, located at

flood insurance claims, which are binding upon the federal government, a WYO insurer is
authorized to use staff adjusters or independent contractors selected and supervised by the
company. The WYO insurer also determines when and how adjusters will be compensated for
their work on flood claims.
WYO insurers typically receive an expense allowance for policies written and claims processed,
and the federal government assumes total financial responsibility for underwriting losses. Insurers
retain 15% of premiums written to cover commissions and salaries of agents and brokers. They
are also reimbursed for marketing, operating, and administrative expenses. The expense ratio of 41
the WYO appears to be about one-third of premiums.
On January 18, 2006, the Collins Center for Public Policy issued a report that raised several
specific concerns involving contracting with private insurance companies for NFIP policy claims
adjusting and servicing. The issue paper indicated (1) WYO insurers had no financial incentive to
adjust claims in the best interest of the NFIP because the federal government pays all claims, and
(2) the existence of conflicts of interest for WYO insurers and their agents when they attempt to
adjust possibly competing claims against flood insurance offered by NFIP, and claims against 42
their own policies for wind or other coverage.
Critics have charged that the NFIP appears to have few systemic checks and balances to ensure
that it is being administered correctly. FEMA has responded to this charge by noting that WYO
insurers are subject to certain standards and oversight as detailed in the NFIP’s “Write Your Own
Program Financial Control Plan Requirements and Procedures Manual.” WYO insurers, for
example, must comply with monthly financial and statistical transaction reporting requirements.
They are subject to a review of operations—claims, underwriting, customer service, marketing,
and litigation activities—every three years to assure that each company is meeting its
performance objectives and adhering to program standards and policies. In addition, WYO
insurers are subject to a “Biennial Claims Audit” every two years, and a “Claims Reinspection
Program” that randomly reviews a percentage of WYO insurers’ claims settlement practices.
Congress may debate the WYO claims adjusting issue given that, under legislation passed in the th
109 Congress, the Department of Homeland Security Inspector General was required to
investigate and report to Congress by April 1, 2007 on whether insurers under the WYO Program
improperly attributed damages from Hurricane Katrina to flooding covered under the National 43
Flood Insurance Program rather than to windstorms covered by such insurers.
Second is the issue of causation in the homeowners’ claims adjustment process—a major source
of insurance-coverage disputes between policyholders and their insurers: i.e., whether a loss was
caused by winds (typically covered under homeowners’ policies) or flooding accompanying the
storm (typically excluded under homeowners’ policies). Causation is the key factor involved in
the water damage exclusion and the “wind versus flood” legal disputes in Louisiana and 44

41 See Collins Center for Public Policy, Issue Paper: National Flood Insurance Program, located at
http://www.fld m/ PressOffice/Do cu ments/Retrieve Do cu ment.asp ?Documen tID=%7 B115A82E4-01B1-44 C6 -
42 Ibid.
43 Department of Homeland Security Appropriations Act of 2007, P.L. 109-295; 120 Stat. 1357.
44 Causation issues may extend beyond the question of windstorm versus flood. For example, damage to buildings may

Homeowners insurance claims are typically paid if the loss is caused by a covered peril or, in the
case of “all risk” (open-perils) policies, a peril that is not expressly excluded. If the loss is caused
by multiple factors (mixed cases), such as wind that is covered and flood that is not, claims
adjustment becomes more challenging. Both wind and flood, for example, might have worked
together to cause the loss; one may have followed directly from the other, or the two may have
arisen independently. In either case, the causal nature of the relationship between the perils has
resulted in legal disputes between policyholders and their insurers. In order to resolve these
mixed cases, the courts have developed various tests for situations in which an excluded peril and
a non-excluded peril contributed to the loss, with the most prominent being the “efficient 45
proximate cause” doctrine and the “concurrent causation” doctrine.
The exact application of causation doctrines—”concurrent causation,” “proximate cause,”
“efficient proximate cause,” and “anti-concurrent causation”—in the context of Hurricane Katrina
claims will depend on the type of insurance policy owned by the claimant, the facts of the claim,
and the case law in the pertinent jurisdiction. Insurance lawyers have observed that every legally
disputed case has unique features with different fact patterns and different levels and types of
coverage that apply to each policyholder.
The next four sections examine causation doctrines or “rules of insurance policy interpretation”
that could be used in understanding the tension between intended meaning in policy language and
interpretation of certain provisions.
Claimants’ attorneys have focused considerable attention on finding ways to overcome decades of
legal precedent supporting the flood exclusion. The principle of concurrent causation holds that if
two causes combine to produce a loss or damage, and one of the two causes is excluded (e.g.,
flood) and the other is covered (e.g., windstorm), the loss will be covered absent policy wording 46
to the contrary. “Proximate cause” and “efficient proximate cause” are variations on “concurrent

be linked to pre-existing structural deficiencies or to outside forces such as vandalism. Mold in buildings may have
been there before any flooding. Yet another issue is damages resulting from the levee break. A key question is: What is
the chain of causation? The breach of a levees built by government authorities would presumably involve two entirely
independent causes that combine to result in loss or damage, but neither set the other in motion. In this case,coverage
under a policy is equally available to an insured whenever an insured risk constitutes simply a concurrent proximate
cause of the injuries.” Policyholders have been able to argue successfully that while a policy might exclude flooding, it
did not exclude the independent concurrent cause of negligence in design or maintenance of levees. In other words,
because the loss resulted from the occurrence of an excluded hazard and a covered hazard, it should be covered.
45 Seth A. Tucker and Ann-Kelley Kemper, “Hurricane Katrina Insurance Coverage Issues, Covington & Burling,
October4, 2005, located at
46 See Doug Simpson’s weblog, Unintended Consequences: Flood Insurance and Exclusions, Proximate and
Concurrent Causation, located at

The proximate cause concept is particularly important in the “chain of causation” arguments
involving wind versus flood claims following a major hurricane. Under the proximate cause
concept, if a policy covers fire, not only is the direct damage by fire covered, but the collateral
smoke damage—as well as damage from the water used by firefighters—is treated as a loss by
fire. The key is that the damage must actually have been caused by the fire and any collateral
damage that did not break the “chain of causation.”
Proximate cause may impose limits on scope of legal liability because determining the cause of a
loss can only be based on what is administratively possible and convenient, i.e., it may not be
determinable with precision.
Given the uncertainty of determining actual “proximate cause,” the efficient proximate doctrine is
used by some courts to allocate losses when damage results from a combination of both covered
and excluded causes. Thus, in the event of multiple causes for a loss, the efficient proximate
cause doctrine allows payment under the policy if the non-excluded cause is the dominant cause
of the loss (i.e., the one that sets the others in motion), notwithstanding that an excluded peril may 47
have contributed to the loss.
Attorneys and the courts frequently utilize the “but for” rule in determining how far back the
chain of causation should go. The rule, as applied to the “wind-v.-flood” coverage dispute, says
that “but for” the particular event (e.g., wind), the loss experienced by the plaintiff would not
have occurred, even when the loss and the covered event are separated by a chain of events that
include floods.
As in any insurance coverage litigation involving whether coverage should be available when
both covered (wind) and excluded (flood) losses are part of a chain of events, there are
differences of opinion on the proper scope of the efficient proximate cause doctrine, particularly
whether the cause is considered a minor or major factor in producing the injury or damage.
In response to the successful use of the concurrent causation arguments and the courts’
reinterpretation of the flood exclusion language that has led to unanticipated exposures, insurers
have sought to draft, file, and get state approval of policy language to make it as clear and
unambiguous as possible that no damage due to flood is covered. Hoping to avoid the unexpected
consequences of future adverse court decisions, insurers sought to implement industry-standard
language designed to cover “all risks” not specifically excluded by the contract language. The
“anti-concurrent causation” doctrine was designed to prevent the theory of concurrent causation
from providing coverage for losses never intended to be covered by standard property insurance
policies. Plaintiffs’ attorneys, however, frequently argue that the exclusions are in violation of
public policy.

47 For expert commentary on causation in insurance contract interpretation, see The Enigma of Causation in Insurance
Contract Interpretation, Kenneth S. Wollner,

Generally speaking, and notwithstanding some federal court decisions to the contrary, the
efficient proximate cause doctrine has been adopted by the highest courts of Mississippi, 48
Louisiana, and Alabama. Courts in these three states have interpreted the doctrine to allow
policyholders to recover for hurricane-related losses where their evidence showed that wind was
the proximate cause of the damage, even if flooding contributed to the loss. To attempt to defend
against the claim, insurers must counter the insured’s evidence with evidence tending to prove
that the proximate and efficient cause was one that falls outside of the coverage of the insurance 49
Recently decided cases, specifically in Mississippi, suggest that the federal courts will enforce the
flood exclusion to the extent the damage is caused by flood, not wind and rain, but will not
enforce anti-concurrent causation language where it would eliminate coverage for otherwise 50
covered damage. In addition, because insurers bear the burden of proving allocation of loss
between wind/rain (covered) and flood (excluded), insurers will likely be held liable for paying 51
all Katrina damage for which the cause of loss cannot be definitely established.
In addition to the application of the efficient proximate cause doctrine to uphold or not uphold
coverage in favor of policyholders, most states and jurisdictions have regulations and rules on
claims handling that provide certain protection to policyholders. Most states have adopted a
version of the National Association of Insurance Commissioners (NAIC) Model Unfair Claims
Practices Act that governs how insurers must deal with claimants. These statutes require
insurance companies to handle claims with good faith and fair dealing. Insurers, for example,
must pay claims according to standards of practice, which is within 30 days after receipt of a
“Proof of Loss”—i.e., a legal document that states the amount the policyholder is claiming under
the policy. In the course of insurance litigation, the interpretation of specific insurance statutory
and regulatory provisions comes into play. Lawsuits frequently arise alleging insurer violation of
specific provisions (e.g., unfair claim settlement practices, delay in settling claims, etc.).

Katrina spawned hundreds of lawsuits against insurance companies,52 most challenging insurers’
reliance on the water-damage exclusion and the anti-concurrent causation language in
homeowners’ policies to deny property damage claims. Among the issues is the extent to which
the anti-concurrent causation language is consistent with or contrary to the settlement law in a

48 Western Assurance Co. v. Hann, 78 So. 232 (Ala. 1971); Glens Falls Ins. Co. of Glens Falls, N.Y. v. Linwood
Elevator, 130 So. 2d 262, 270 (Miss. 1961); Evans Plantation, Inc. V. Yorkshire Ins. Co, 58 So. 2d 797, 798 (Miss.
1952); Roach-Strayhan-Holland Post No. 20, American Legion Club, Inc. v. Continental Ins. Co. of N.Y., 112 So. 2d
680 (La. 1959).
49 See, e.g., Broussard v. State Farm Fire and Casualty Company, Civil Action No. 1:06cv6-LTS-RHW (S.D. Miss.
1/11/2007) (on cross-motions for judgment as a matter of law).
50 E.g., Leonard v. Nationwide Mutual Ins. Co., 438 F. Supp. 2d 684 (S.D. Miss. 2006).
51 Broussard, supra.
52 Daniel Hays and Susanne Sclafane, “Mississippi Negotiating Katrina Settlements: Deal With State Farm Reportedly
Near, National Underwriter: Property & Casualty, January 15, 2007.

given state. This section provides a brief, narrative summary of major legal activities related to
Katrina insurance claims disputes.
Policyholders typically use the “valued policy” statutes to gain coverage in cases where both
wind and water caused damage to the property. Valued policy statutes require insurers to
determine the value of the property being insured at the time a policy is written and to pay that
full value to the insured when there is a total loss caused by a covered peril. The insured is not
required to prove the value of the damaged property. A building is considered a total loss when
the necessary repair costs are more than 50% of the value of the building.
Both Louisiana and Mississippi have enacted “valued policy” statutes as follows:
The Louisiana Valued Policy statute provides that:
Under any fire insurance policy insuring inanimate, immovable property in this state, if the
insurer places a valuation upon the covered property and uses such valuation for purposes of
determining the premium charge to be made under the policy, in the case of total loss the
insurer shall compute and indemnify or compensate any covered loss of, or damage to, such
property which occurs during the term of the policy at such valuation without deduction or
offset, unless a different method is to be used in the computation of loss, in which latter case,
the policy, and any application therefore, shall set forth in type of equal size, the actual
method of such loss computation by the insurer.
The Mississippi Valued Policy statute provides that:
No insurance company shall knowingly issue any fire insurance policy upon property within
this state for an amount which, taken with any existing insurance thereon, exceeds a fair
value of the property, nor for a longer term than five years. When buildings and structures
are insured against loss by fire and, situated within the state, are totally destroyed by fire, the
company shall not be permitted to deny that the building or structures insured were worth at
the time of the issuance of the policy the full value upon which the insurance is calculated
and the measure of damages shall be the amount for which the building and structures were
Mississippi has an instructive precedent regarding the water damage exclusion. Under Mississippi
common law, where there is damage caused by both wind and rain (covered loss) and flood
(excluded loss), the amount due under the policy will generally depend on the proximate and
efficient cause of the damage (i.e., hurricane wind), even if other “non” covered causes also 53
contributed to the loss.
A Florida case found that the state’s Valued policy statute did, in fact, require payment of the face
amount of the wind-insurance policy because the insured property was damaged—at least in 54
part—by a covered peril (wind). The special concurrence in Mierzwa would have required that
the “proximate cause” of the damage be the covered peril (wind) (and not merely, as the majority
held, that it be some part of the cause of the damage), but agreed with the majority result because

53 See Grace v. Lititz Mutual Insurance Co. 257 So. 2d 217 (Miss. 1972).
54 Mierzwa v. Florida Windstorm Underwriting Association, 877 So.2d 774 (Fla. App. 2004).

the concurrer found such “proximate cause.”55 The Florida statute, however, was amended after
the Mierzwa decision to specifically allow for the pro rating of damages caused by both a covered 56
and a non-covered peril except “if the covered perils alone would have caused the total loss.”
Several cases from Louisiana have been filed, presumably seeking to rely on Mierzwa, attempting
to get insurers to cover their damages—notwithstanding that the Louisiana valued policy statute
is directed specifically at fire insurance policies.
A suit filed in Mississippi Chancery Court by the Mississippi Attorney General in September
2005 against State Farm Fire and Casualty Company and other insurers sought to enforce the
insurance policies issued by defendant insurers. Although the defendants, who maintained that the
policies exclude coverage of damages caused by flooding, attempted to have the case removed to
federal court on the ground that it involved litigation concerning a federal program (the insurers
are WYO issuers of flood insurance policies pursuant to the NFIP), the United States District
Court for the Southern District of Mississippi, on reconsideration of an earlier Order to remand
the case to the state court, affirmed the Order, stating
[Defendants] have attempted to create federal jurisdiction out of what is essentially ... ‘the
interpretation of the terms of private [homeowners] insurance policies, traditionally a 57
function of state law.
State Farm agreed to a proposed settlement of the case, in conjunction with a pending, private, 58
proposed, class-action suit, and the settlement was concluded on January 23, 2007. Not only 59
was the proposed class denied, the court rejected the settlement on several grounds:
The proposed settlement agreement establishes certain absolute limits on class members
potential recoveries that may be inconsistent with my prior rulings. ... The agreement
[pursuant to which the insurers liability is limited by the amount of recovery under separate
flood-damage policies] does not provide for any exception for situations in which the fair
market value or the actual cash value of the insured property is equal to or greater than the 60
combined limits of flood coverage limits and the coverage limits in the State Farm policies.
In addition, the court was bothered by State Farm’s “indirect control” over the claims handling
procedure because the appointment by State Farm of a Special Master to oversee the process 61
would impinge on a function “that is exclusively within the prerogative of the Court.”

55 877 So. 2d at 781-2.
56 See Wests F.S.A. § 627.702(1)(b), effective 6/1/2005.
57 Hood v. Mississippi Farm Bureau Insurance Company, 2006 WL 3802170 (S.D. Miss. 2006), quoting from the
earlier Order issued by Judge Tom S. Lee.
58 Woullard v. State Farm Fire and Casualty Co., Docket No. 1:06cv1057 (S.D. Miss.). Available at
59The plaintiffs have alleged that ‘many hundreds, if not thousands, of individuals and/or entities have asserted claims
or have potential claims’ against State Farm. Neither the plaintiffs nor State Farm has given the Court any information
from which the Court can determine with any reasonable degree of certainty how many policyholders are within the
proposed class or how many policyholders have each of the eleven types of policies identified.Id.
60 Id.
61 Id.

Moreover, the complexity of the claims procedure would have prevented the effective
participation of many claimants. The court was also uncomfortable with “sending a large number
of policyholders into the process of binding arbitration when none of these individuals have ever
agreed to participate in that procedure.” These facts were considerable obstacles that led to the 62
court’s rejection of the settlement.
There are other, private, ongoing cases in both Mississippi and Louisiana challenging insurers’
interpretations of policies relied upon for protection against hurricane-related damage. A list of 63
the opinions or orders in those cases is available on the website of the U.S. District Court for the 64
Southern District of Mississippi, as the cases have been consolidated on the docket of that court.
All of the cases involve some permutation of the wind v. water/flood equation.
The Attorney General of Mississippi, Jim Hood, began a criminal investigation into State Farm’s
handling of Katrina claims. A grand jury was seated to hear the charges that focused on the wind
v. water debate. Using documents provided by whistleblowers who worked for an insurer, and the
state’s Unfair Claims Practices statute, Jim Hood alleged insurers defrauded policyholders by
manipulating engineering reports to deny claims. On January 23, 2007, as part of the settlement
with State Farm on the class action litigation, Hood ended the criminal investigation, opting
instead to handle the matter in civil court and in Congress. He supports congressional oversight 65
and investigation on this matter and national insurance reform.

Three sets of bills—H.R. 920/H.R. 3121/S. 2310; H.R. 537/S. 292, and H.R. 1081/S. 618—have th
been introduced in the 110 Congress to address post-Katrina insurance coverage issues. These
measures would:(1) enhance insurance reform of the NFIP (H.R. 920/H.R. 3121/S. 2310); (2)
establish a bipartisan commission to study the Gulf states’ insurance market in Katrina’s
aftermath and make recommendations to Congress regarding the availability of insurance for
catastrophic risks (H.R. 537/S. 292); and (3) repeal the insurance industry’s limited exemption
from federal antitrust laws to make the industry more competitive (H.R. 1081/S. 618).
On February 8, 2007, Representative Gene Taylor introduced the Multiple Peril Insurance Act of
2007 (H.R. 920), which would create an all-peril policy that covers both wind- and water-related
damages for both homeowners and small businesses under the NFIP. The legislation seeks to
expand coverage offered by the NFIP from flood only policies to include flood and wind perils.
H.R. 920 is also designed to reduce complexity in claims adjusting associated with wind and
flood loss segregation, enhance wind insurance availability, and reduce prices.

62 Id.
63 Descriptions of the litigation of those cases is contemplated in a forthcoming addendum to this report.
64 See
65 Michael Kunzelman, “Ex-State Farm Adjusters Tell Mississippi Grand Jury of Katrina Claims, Associated Press,
January 23, 2007, located at

On July 19, 2007, Representative Maxine Waters introduced H.R. 3121 to restore the financial
solvency of the national flood insurance program. Chairman Barney Frank had introduced H.R.
1682, an earlier version of H.R. 3121, on March 26, 2007. H.R. 3121 would make the NFIP
satisfy traditional criteria for actuarial soundness by phasing out discounted premiums. H.R. 3121
would also: (1) allow the Federal Emergency Management Agency (FEMA) to increase flood
policy rates by 15% a year, up from 10%; (2) increase the program’s borrowing authority to $21.5
billion from $20.8 billion; (3) raise civil penalties on federally regulated lenders who fail to
enforce mandatory purchase of flood insurance for mortgage holders; (4) increase program
participation incentives; (5) add coverage for wind as well as water damage, as outlined in H.R.
920; and (6) encourage the revisions to flood maps. The bill passed the full House on September
27, 2007. On November 1, 2007, Senator Christopher J. Dodd introduced S. 2284, a flood
insurance reform bill designed to increase the amount of premiums collected reduce the cost of
expected claims under the NFIP. S. 2284 would also forgive the program’s outstanding debt to the
On February 16, 2007, Senator Patrick Leahy and Representative Peter A. DeFazio introduced the
Insurance Industry Competition Acts of 2007 (S. 618/H.R. 1081), which would amend the
McCarran-Ferguson Act of 1945 to give the Federal Trade Commission and the Justice
Department oversight over ensuring that insurers comply with federal antitrust laws. The
McCarran Ferguson Act allows collaborative industry practices like the development of
standardized policy language which makes it easy for consumers to compare policies and prices.
On January 12, 2007, Senator Bill Nelson and Representative Kendrick B. Meek introduced the
Commission on Catastrophe Disaster Risk and Insurance Acts of 2007 (S. 292/H.R. 537) to create
a federal bipartisan commission to study catastrophe insurance markets in Katrina’s aftermath and
make recommendations to Congress regarding the availability of insurance for catastrophic risks.
The commission would establish a forum for both the insurance and consumers to address post-
Katrina insurance issues.
Finally, the 109th Congress enacted the Department of Homeland Security Appropriations Act of
2007 that included a provision that directed the DHS Inspector general to investigate and report to
Congress on whether insurers under the NFIP’s Write-your-Own program improperly attributed
damages from Hurricane Katrina to flooding covered under the National Flood Insurance
Program rather than to windstorms covered by such insurers.
The section of the DHS Appropriations Act reads as follows:
...the Department of Homeland Security Inspector General shall investigate whether, and to
what extent, in adjusting and settling claims resulting from Hurricane Katrina, insurers
making flood insurance coverage available under the Write-your-own program pursuant to
section 1345 of the National Flood Insurance Act of 1968 (42 U.S.C. 4081) and subpart C of
part 62 of title 44, Code of Federal Regulations, improperly attributed damages from such
hurricane to flooding covered under the insurance coverage provided under the national
flood insurance program rather than to windstorms covered under coverage provide by such
insurers or by windstorm insurance pools in which such insurers participated...the
Department of Homeland Security Inspector General shall submit a report to Congress not 66
later than April 1, 2007, setting forth the conclusions of such investigation.

66 P.L. 109-295, 120 Stat. 1357.

Several concluding observations could be made.
• One major problem that has been identified is that the NFIP flood coverage and
the private market wind coverage are provided under separate coverage forms,
many times by separate insurers and often through separate distribution
mechanisms. Congress might choose to focus on the identified problem: flood
and wind coverages do not intersect. The NFIP’s WYO program works today
because the flood policy is typically sold by the same personal lines agent and
through the same personal lines insurer that offer the basic property policy. The
challenge for policymakers and insurers is to develop better coordination
between flood and wind coverage.
• A major problem with the current system of flood risk mapping is the
incorporation of the latest information on risk. When there is evidence that risk
levels are rising or that risk was previously underestimated, it can be difficult to
get the appropriate adjustments approved. As an illustration, climatologists have
observed that the nation is in a period of higher hurricane activity and rising sea
levels. The NFIP’s coastal storm surge flood zones at any return period may
extend further inland than are shown on the official FEMA flood maps. As a
result, the construction of buildings at dangerously low elevations will continue
to be permitted. In 2006, FEMA announced it was in the process of issuing new
flood maps for the City of New Orleans and areas of Mississippi. Until such time
as these maps are available, FEMA has issued Advisory Base Flood Elevations
(BFEs), which direct areas “protected by levees to elevate substantially damaged
homes and businesses to 3 feet above the highest adjacent existing grade on site
or the current BFE on the flood insurance rate map, whichever is higher.”
• NFIP has a low level of market penetration. Despite a requirement that many
policyholders purchase NFIP coverage, a much lower than expected number do.
Many of those that claim “wind / flood” issues appear to be attacking the private
insurance policy’s flood exclusion because the policyholder did not purchase
NFIP coverage. Congress faces relatively low levels of participation in the NFIP.
• Is there a need for a national catastrophe insurance solution? The media has made
references to insurance affordability and availability issues in many parts of the
nation. One solution offered is to have the federal government assume wind risk
on a nationwide basis. Hurricane Katrina resulted in record-setting insured and
uninsured property damages, sharply higher insurance rates, and insurance
availability problems for homeowners and businesses in coastal areas. Because of
rising cost or dissatisfaction with insurers, a growing number of homeowners and
small businesses in Louisiana and Mississippi are reducing insurance coverage,
or dropping it altogether. State-sponsored catastrophe insurance pools suffered
record losses and experienced major deficits following Hurricanes Katrina and
Rita. These indications of financial distress might be considered precursors to
potential market failure. Some question, however, whether federal disaster
legislation is needed if private sector wind insurance coverage is sufficient in
non-hurricane-prone areas.

The issue before Congress is whether wind insurance market failure exists
outside of hurricane-prone Gulf and Atlantic coast areas. Insurance market failure
could justify federal government intervention. One possible precedent can be
seen in flood insurance. In the decades of the 1920s through the 1960s, private
insurers were unwilling and unable to provide flood coverage. In response, the
federal government created the National Flood Insurance Program (NFIP) in
1968. While the private insurance sector failed to provide flood coverage, these
same firms have always been willing and able to offer wind coverage to property
owners residing outside of hurricane-prone areas. Some question whether or not
states need to disband well functioning system as part of a nationalization of
wind insurance. If the federal government offers optional wind insurance
coverage in NFIP-designated communities not subject to hurricane risk, and
where private insurers now offer coverage at affordable prices, the availability of
federal wind insurance coverage could arguably compete directly with the private
insurance industry. The insurance availability and affordability problem seems
localized, but some have proposed a national solution. In order to avoid market
competition, a national wind insurance program could target areas where the
insurance regulator approved the wind damage policy exclusion peril and there
are signs of insurance market failure.
• Government insurance only appears cheaper than private insurance. Federal
insurance programs like any of the state catastrophe funds or federal insurance
programs appear cheaper up front than they really are. Because government can
use taxpayer capital without compensating the taxpayer, government can provide
an up-front cost of insurance that is “actuarially sound” for much less than
private insurers. This is how the NFIP has functioned to date. Now, there is a call
on the taxpayer’s capital in the form of a $22 billion deficit. Analysts ask: How
was the taxpayer compensated for taking that risk? If NFIP was a private
company, then it would have had to acquire capital from investors and/or
reinsurers—and pay them for that capital. Government can implicitly tax the
taxpayer pre-event by shifting the risk to large future post-event expenditures,
thereby denying taxpayers compensation for that risk. In this way, government
can make insurance appear cheap on the front end but with enormous post-event
Rawle O. King
Analyst in Financial Economics and Risk
Assessment, 7-5975