Hurricanes and Disaster Risk Financing Through Insurance: Challenges and Policy Options
Order Code RL32825
Hurricanes and Disaster Risk Financing Through
Insurance: Challenges and Policy Options
Updated January 31, 2008
Rawle O. King
Analyst in Financial Economics and Risk Assessment
Government and Finance Division
Hurricanes and Disaster Risk Financing Through
Insurance: Challenges and Policy Options
Summary
The U.S. Atlantic and Gulf of Mexico coastal states, Hawaii, Puerto Rico, and
the U.S. Virgin Island are exposed to relatively high levels of risk from hurricanes
and tropical storms. To address the financial and economic effects of such risks,
residents and business owners have relied on private insurance, state-sponsored
insurance pools, and/or federal emergency disaster assistance to manage their natural
hazard risk.
The 2005 hurricane season was devastating to residents and businesses in the
coastal high-hazard areas of the Gulf Coast and New Orleans. Although insured
catastrophe losses in 2005 totaled $61.2 billion, the industry had the financial
resources to pay all insured claims without threatening its solvency and claims-
paying ability. . Hurricane Katrina losses was $41.1 billion.
Insurance industry participants, legislators, and policymakers learned a great deal
from both Hurricane Andrew in 1992 and the four major hurricanes during the 2004
season, and they took specific actions that had the effect of minimizing the impact of
the 2005 hurricane season. Nevertheless, most disaster experts and policymakers did
not anticipate nor were they prepared for the magnitude of flood damage and the
subsequent number of flood claims filed in the wake of Hurricane Katrina. As a
result, the U.S. Congress was called upon to consider major revisions to the National
Flood Insurance Program (NFIP). On September 27, 2007, the full House approved
H.R. 3121, the Flood Insurance Reform and Modernization Act of 2007, to restore the
financial solvency of the program. On November 1, 2007, Senator Christopher Dodd
introduced S. 2284 to reform the flood insurance program while forgiving the
estimated $17.5 billion in debt owed to the U.S. Treasury, as of December 31, 2007.
With respect to the broader issues of managing and financing catastrophe risk,
members of the 110th Congress might also focus attention on the long-term budgetary
implications of disaster recovery expenses incurred by the federal government, and
finding ways to expand private-sector capacity for insuring disaster losses. Previous
Congresses responded to insurers’ concerns by considering legislation to create a
federal catastrophe reinsurance program for residential property.
This report examines the role of insurance in financing disaster risk and the
changes implemented by insurers and legislators that helped to minimize market
disruptions following the 2005 and 2004 hurricane seasons. After reviewing the
congressional interest in financing catastrophe risk and summarizing the results of
the last two hurricane seasons, the report describes lessons learned, the insurance
market’s response to hurricanes, and existing mechanisms for insuring hurricane
losses. The concluding two sections analyze issue and policy options as well as
future challenges that policymakers in the 110th Congress face.
This report will be updated as events warrant.
Contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Financing Catastrophic Risk With Insurance . . . . . . . . . . . . . . . . . . . . . . . . . 2
Congressional Interest In Financing Catastrophic Risk . . . . . . . . . . . . . . . . . . . . . 3
The 2005 Hurricane Season . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
The 2004 Hurricane Season . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Insurance Lessons Learned from Hurricane Andrew (1992) . . . . . . . . . . . . . . . . 12
Insurance Market Response to Past Hurricanes . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Hurricane Insurance Deductibles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Capital Market Instruments Linked to Catastrophe Risk . . . . . . . . . . . . . . . 16
Building Codes and Construction Standards . . . . . . . . . . . . . . . . . . . . . . . . 16
Catastrophe Modeling and Insurance Underwriting . . . . . . . . . . . . . . . . . . 17
Transferring Risk Through Insurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Federal Flood Insurance Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
State-Sponsored Lost-Sharing Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . 20
Fair Plans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Beach and Windstorm Insurance Plans . . . . . . . . . . . . . . . . . . . . . . . . 21
Market Assistance Plans (MAP) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Surplus Lines Insurance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Challenges for the 110th Congress and Beyond . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Population Growth and Coastal Development . . . . . . . . . . . . . . . . . . . . . . . 25
Rising Property Values in Coastal Areas . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Climatological and Environmental Changes . . . . . . . . . . . . . . . . . . . . . . . . 27
Issues and Policy Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
List of Tables
Table 1. Value of Insured Coastal Properties Vulnerable to Hurricanes By State,
2004 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Table 2. Total U.S. Insured Losses and Federal Outlays for Uninsured Losses
from Major Disasters: 1995-2005 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Table 3. Ten Most Costly Catastrophes in the United States . . . . . . . . . . . . . . . . 9
Hurricanes and Disaster Risk Financing
Through Insurance: Challenges
and Policy Options
Introduction
The eighteen states along the U.S. Gulf and Atlantic coast,1 extending from
Texas to Maine, along with Hawaii, Puerto Rico and the U.S. Virgin Islands, are at
relatively high risk from hurricanes and tropical storms.2 Hurricanes and tropical
storms typically produce violent winds, heavy rains, and storm surges that result in
flooding, coastal erosion, and ecological damage. When they strike in populated,
commercial, or industrial areas, hurricanes and tropical storms can cause dozens of
deaths and billions of dollars in both direct costs (e.g., loss of capital stock and
investments) and indirect costs (e.g., disruption of economic activity, including loss
of income, employment and services).3
Hurricanes are normally described as being in one of five categories, depending
on their wind velocity. Category one hurricanes have winds of 74 to 95 miles per
hour, category two hurricanes have winds of 96 to 110 miles per hour, category three
hurricanes have winds of 111 to 130 miles per hour, category four hurricanes have
winds of 131 to 155 miles per hour, and category five hurricanes have winds greater
than 155 miles per hour.
Lessons learned from the 2005 and 2004 hurricane seasons have led the 110th
Congress to focus attention on the mounting cost of federal outlays for disaster
assistance involving hurricanes, and deciding whether and how the federal
government could improve the nation’s ability to finance the losses created by these
events. Insurers, legislators and policymakers learned a great deal from the
devastation caused by Hurricane Andrew in 1992, especially in the areas of pre-
disaster mitigation and the financing of catastrophic risk, and actions they took
served to minimize market disruption following both the devastating 2005 and 2004
1 These 18 states are: Alabama, Connecticut, Delaware, Florida, Georgia, Louisiana,
Maryland, Massachusetts, Mississippi, Maine, New Hampshire, New Jersey, New York,
North Carolina, South Carolina, Texas, Vermont, Virginia.
2 Hurricanes are formed in the North Atlantic, Caribbean Ocean, Gulf of Mexico, and the
Pacific Coast of Mexico. The greatest likelihood of a hurricane striking land areas is along
the Gulf Coast and the Southeastern Seaboard, as well as Hawaii. Some hurricanes have
struck central Pennsylvania and the coast of New Jersey, New York, Maryland, and New
England.
3 Rachel A. Davidson, and Kelly B. Lambert, “Comparing the Hurricane Disaster Risk of
U.S. Coastal Counties,” Natural Hazards Review, August 2001, p. 132.
CRS-2
hurricane seasons. However, the short-term insurance solutions designed to finance
loss caused by a small (category one) or moderate hurricane (category two, three or
four) will not work for a catastrophic hurricane (category five) because state pools
lack the financial capacity for financing events of such magnitude.
Prior to the beginning of the 110th Congress, some Members of Congress had
begun to rethink federal disaster policy, particularly with respect to the financing of
catastrophic risk and the unwillingness or inability of insurers to provide property
insurance coverage to the extent sought and needed. An important issue these
Members grappled with was deciding how to reconcile the possible roles for the
public and private sectors in disaster risk financing and risk reduction. Several
questions arose: What has been the experience of using financial tools, such as
insurance and other financial services, to reduce disaster risk? What challenges and
opportunities exist for disaster risk transfer and risk reduction schemes? And lastly,
what concrete steps must be taken, and by whom, to form partnerships between the
public and private sectors to use insurance and other financial services for disaster
risk reduction?
In preparation for continued debate on financing disaster risks in the 110th
Congress, this report examines the role of insurance in financing disaster risk and the
changes implemented by insurers and state legislators that helped to minimize market
disruptions following the 2005 and 2004 hurricane seasons. After reviewing the
congressional interest in financing catastrophe risk and summarizing the results of
the last two hurricane seasons, the next three sections describe lessons learned,
insurance market’s response to hurricanes, and existing mechanisms for insuring
hurricane losses. The concluding sections analyze issues and policy options as well
as future challenges that policymakers in the 110th Congress face.
Financing Catastrophic Risk With Insurance
Individuals and policymakers have two options to reduce losses from disasters:
pre-disaster mitigation that reduces physical /environmental vulnerabilities and risk
financing designed to reduce financial vulnerabilities. The first step in the disaster
management framework is to mitigate damages from disasters. The residual
economic risk can then be managed with risk financing strategies. Financing is thus
an integral part of managing disaster risk; it would not be feasible to quickly
reconstruct the damaged property and infrastructure, and also to restore the livelihood
of the affected persons without adequate financial arrangements.
Insurance is the primary method of financing natural disaster losses in the U.S.
Risk financing with insurance avoids the time lag that is associated with post-disaster
assistance or financing. Insurers are able to assess damages and reimburse disaster
victims for financial losses. In providing insurance coverage, an insurer will agree
to assume a portion of the policyholder’s disaster risk exposure in exchange for a
premium. From this premium payment, the insurer sets aside loss reserves to pay
expected claims and build up capital reserves to “buffer” against the risk of insurer
insolvency from low-probability, high-cost events. Insurance companies supplement
CRS-3
this arrangement by purchasing reinsurance from a reinsurance company so that
losses from a catastrophic event are spread worldwide.4
Most insurance experts agree that although primary insurers and traditional
reinsurers could absorb the loss shock from a moderate (category two, three, or four)
hurricane (e.g., less than $50 billion in insured losses), their financial capacity may
not be adequate to cope with a catastrophic (category five) hurricane. Estimates of
the probable maximum losses (PMLs) from a catastrophic hurricane striking the U.S.
range up to $100 billion, and this figure could be even higher depending on the
location, time and intensity of the event. The PML loss from a Category 5 hurricane
directly hitting a densely populated area along the Gulf and Atlantic Coast (e.g., the
Miami-Ft. Lauderdale area) could exceed the total capacity (policyholder surplus) of
the U.S. insurance industry.5 The policyholders surplus of the entire property and
casualty insurance industry stood at approximately $521.6 billion at the end of 2007.6
Only a fraction of this industry-wide total surplus amount would be available to
compensate victims of a hurricane. Insurers must rely on this same limited pool of
capital to pay for other potentially catastrophic and unpredictable risks, such as
terrorism, mold, and medical malpractice and asbestos liability claims. Insurers may
have to liquidate bonds and other financial assets in order to pay claims, triggering
an adverse impact on U.S. financial markets.7
Congressional Interest In Financing
Catastrophic Risk
The property and casualty insurance industry will be able to pay all claims from
the devastating 2005 hurricane season without jeopardizing the solvency and claims-
paying ability of the industry as a whole. In the absence of an efficient and effective
catastrophe insurance market — one that provides property insurance at reasonable
rates to residents and business owners and spreads catastrophe risk globally — the
government often becomes the de facto financier of disaster recovery efforts through
4 Reinsurance provides considerable protection to the primary insurer by: (1) limiting that
insurer’s loss exposure to levels commensurate with their net assets; (2) reducing the wide
swings in profit and loss margins inherent to the insurance business; (3) protecting against
catastrophic loss; and (4) increasing capacity or the dollar amount of risk an insurer can
prudently assume, based on its surplus and the nature of the business written.
5 David J. Cummins, Neil A. Doherty and Anita Lo, “Can Insurers Pay for the ‘Big One’?
Measuring the Capacity of an Insurance Market to Respond to Catastrophic Losses,”
Journal of Banking and Finance, vol. 26, no. 2, p. 557.
6 Policyholders surplus refers to “net worth” or “owners’ equity” in other industries. It is
a measure of the capacity of insurers to underwrite policies, and it must increase to meet the
demands of a growing U.S. economy and claims resulting from hurricanes and other natural
hazards.
7 Ross J. Davidson Jr., “Working Toward a Comprehensive National Strategy for Funding
Catastrophe Exposures,” Journal of Insurance Regulation, vol. 7, no. 2, Winter 1998, p.
134.
CRS-4
existing government disaster response and recovery programs and the “traditional”
appropriation process.
As the frequency and severity of natural disasters increase, state governments
have used their taxing and bonding capacities to establish public/private insurance
pools to engage in catastrophe risk management and financing. Recent catastrophes,
however, have strained these pools and most insurance market analysts would agree
the private sector insurance industry has enormous capacity to provide catastrophe
risk insurance, but that coverage is limited. As demonstrated by the 2004 and 2005
hurricane seasons, the industry can readily handle a series of events with insured
damages above $60 billion. Some reinsurers experts and insurance analysts insist the
insurance industry can handle a single insured event approaching $100 billion in
claims.
Hurricane Katrina has focused public debate on whether to implement a
comprehensive (ex-ante or post-disaster financing) solution to the problems
presented by natural catastrophe exposure. In an environment of budgetary deficits
and spending constraints, policymakers are seeking to find ways to confront the
excessive and inequitable reliance upon federal disaster relief. The 110th Congress
might be asked to consider proposals which would permit federal reinsurance payouts
for events over the level of exposure that the private sector can adequately and
appropriately insure.
Table 1, which shows the value of insured coastal properties in the 18 states
along the U.S. Gulf and Atlantic coasts, indicates that $6.8 trillion out of $19 trillion
in insured property is vulnerable to hurricanes. The nation realizes this risk when
hurricanes strike and individuals, businesses, and communities suffer, while
American taxpayers, through the federal government, bear the costs associated with
indemnifying uninsured victims of natural disasters and rebuilding critical
infrastructure.
CRS-5
Table 1. Value of Insured Coastal Properties Vulnerable to
Hurricanes By State, 2004
($ Billion)
Coastal as a
State
Coastal
Total Exposurea
Percent of Total
Florida
$1,937.4
$2,443.5
79%
New York
1,901.6
3,123.6
61
Texas
740.0
2,895.3
26
Massachusetts
662.4
1,223.0
54
New Jersey
505.8
1,504.8
34
Connecticut
404.9
641.3
63
Louisiana
209.3
551.7
38
South Carolina
148.8
581.2
26
Virginia
129.7
1,140.2
11
Maine
117.2
202.4
58
North Carolina
105.3
1,189.3
9
Alabama
75.9
631.3
12
Georgia
73.0
1,235.7
6
Delaware
46.4
140.1
33
New Hampshire
45.6
196.0
23
Mississippi
44.7
331.4
13
Rhode Island
43.8
156.6
28
Maryland
12.1
853.6
1
Coastal States
$6,863.0
$19,041.1
36
Source: AIR Worldwide Corporation.
a. Exposure is the total amount of insured property in the state.
Table 2 shows that while the frequency of catastrophic events in the last few
years is less compared with earlier years, insured losses have increased significantly.
The increasing magnitude of both insured and uninsured losses from natural disasters
represent an ongoing challenge for governments and the private sector. Catastrophes
result in large government outlays for disaster assistance and they place a financial
strain on private disaster insurance/reinsurance markets. The federal government
alone, facing fiscal constraints to cover the losses to the private sector, will find it
challenging to meet long-term disaster-related spending. Further, insurers have been
and will continue to be reluctant to cover properties in high-risk areas because of
high long-run costs (which translates into high prices for disaster insurance) and low
demand for disaster insurance.8 To make insurance available and affordable, state
governments have created public/private partnerships (state-sponsored catastrophe
funds) to provide catastrophe insurance or reinsurance coverage at subsidized rates.
8 The high long-run costs and low demand for disaster insurance results from insurers having
to hold huge amounts of capital to pay claims resulting from rare but potentially large
catastrophe losses, and the limited willingness of many consumers to pay risk-based
premiums for disaster insurance, respectively.
CRS-6
Table 2. Total U.S. Insured Losses and Federal Outlays for
Uninsured Losses from Major Disasters: 1995-2005a
($ millions)
Insured Losses
Uninsured Losses
Insured
Total b
Number
Number
Dollars
Losses
Appropriations
of
of Claims
When
in 2004
(available
Federal
Year
Events
(Millions)
Occurred
Dollars
funds)
Outlaysc
1995
34
2.7
$8,310
$10,033
$4,235
$2,492
1996
41
3.9
7,375
8,649
4,042
2,581
1997
25
1.6
2,600
2,981
5,248
2,898
1998
37
3.5
10,070
11,367
2,155
2,242
1999
27
3.3
8,321
9,190
2,597
4,149
2000
24
1.4
4,600
4,915
3,019
2,853
2001
20
1.6
26,548
27,582
6,249
3,413
2002
25
1.8
5,850
5,932
12,677
4,114
2003
21
2.6
12,885
12,885
2,255
8,761
2004
22
3.4
27,490
27,490
2,068
3,082
2005
24
4.3
56,779
56,779
70,542
N/A
Source: Insurance Services Office, Inc., Jersey City, New Jersey
Note: NA = not applicable.
a. The definition of a catastrophe changed in 1996. Beginning in 1997 the catastrophe definition was
raised from $5 million to $25 million in insured damage. This change might explain why the
number of recorded catastrophes and the aggregate losses attributed to catastrophes on average
is lower than in earlier years. The figures for appropriations and outlays in the last two columns
are different because Congress appropriates funds to make it available, but the actual amounts
spent could be different.
b. Total appropriations into the Disaster Relief Fund. Figures are in 2002 constant dollars. The data
in this column comes from: CRS Report RL32242, Emergency Management Funding for the
Department of Homeland Security: Information and Issues for FY 2005, by Keith Bea.
c. These figures, which are in 2002 constant dollars, come from: CRS Report RL33053, Federal
Stafford Act Disaster Assistance: Presidential Declaration, Eligible Activity Funding, by Keith
Bea.
The last two columns in Table 2 show total appropriations into and outlays from
the Disaster Relief Fund (DRF) for federal disaster assistance to help individuals,
families, state and local governments, and certain nonprofit organizations affected
by severe disasters. Average annual federal outlays exceeded $3.6 billion since 1995
because of significant hurricanes (Andrew and Inniki FY1992), earthquakes
(Northridge in FY1994), floods (Midwest floods of 1993, Red River Floods of 1995)
the terrorists attacks of September 11, 2001, and the sequence of four major
hurricanes in 2004.9 For purposes of illustration, prior to FY1989, outlays from the
DRF averaged $568 million, and on only two occasions exceeded $1 billion.10
9 CRS Report RL33053, Federal Stafford Act Disaster Assistance: Presidential Declaration,
Eligible Activity Funding, by Keith Bea.
10 Ibid, p. 7.
CRS-7
Most disaster experts and policymakers did not anticipate and therefore were
unprepared for the magnitude of flood damage, and the subsequent number of flood
claims filed, in the wake of Hurricane Katrina. Given the magnitude of Katrina-
related damages and their impact on the NFIP, the 110th Congress might be called
upon to consider ways to overhaul the program.
With respect to the broader issue of managing and financing catastrophe risk,
the 110th Congress might choose to focus attention on the long-term budgetary
implications of disaster recovery expenses incurred by the federal government and
finding ways to expand private-sector capacity for insuring disaster losses. The last
time Congress took a critical examination of the federal disaster policy was in 1998.11
This is likely to occur in 2008 at the same time that the property insurance industry
seeks some type of federal assistance in reducing their catastrophe exposure.
Ironically, the insurance industry has historically opposed federal intervention in the
insurance marketplace. But, faced with new terrorism risk following the September
11, 2001, terrorists attacks, and the recognition of a possible mega-catastrophic
hurricane far more devastating than Hurricanes Katrina, Rita, and Wilma, some
insurers and their trade associations have begun to rethink their support for federal
involvement in disaster insurance markets.
Previous Congresses responded to insurers’ concerns about a mega-catastrophe
that threatens the solvency and claim-paying ability of the insurance industry by
considering legislation to create a federal catastrophe reinsurance program for
residential property.12 Despite broad bipartisan support for addressing America’s
exposure to natural disasters, the full Congress did not approve the creation of a
federal reinsurance program until the enactment of the Terrorism Risk Insurance Act
(TRIA) of 2002.13 TRIA provides a temporary federal reinsurance backstop for
future terrorist acts once a high insurance-industry loss is sustained. The law, which
was scheduled to expire on December 31, 2007, was extended through the end of
2014.14
All federal disaster insurance bills have one thing in common: they seek to
improve the nation’s ability to finance catastrophe risk through insurance as opposed
to increased direct spending for federal disaster assistance. Their justification is
based on the argument that such initiatives will: (1) enhance the current catastrophe
funding system; (2) make property insurance more available and affordable in high-
risk areas; (3) promote the funding of research studies (i.e., earthquake science,
actuarial science, economics, and finance) on disaster insurance issues; and (4)
expand our knowledge and understanding of the scientific and financial aspects of
11 See U.S. Congress, House Committee on the Budget, Task Force on Budget Process,
Budgetary Treatment of Emergencies, hearing, 105th Cong., 2nd sess., June 23, 1998
Washington: GPO,
12 Elliott Mitter, “Alternative National Earthquake Insurance Programs,” Earthquake
Spectrum, August 1991, vol. 7, no. 3, p. 757.
13 P.L. 107-297; 116 Stat 2322.
14 CRS Report RL34219, Terrorism Risk Insurance Legislation in 2007: Issue Summary and
Side-by-Side, by Baird Webel.
CRS-8
natural hazards. Professor Howard Kunreuther at the University of Pennsylvania has
suggested that improvements in the scientific and financial areas are thought to be
important because of the urgency in finding ways to predict the probability and
magnitude of future natural hazards, plan for the necessary funding for disaster
recovery, and devise the optimum allocation of resources after the event in order to
promote speedy economic recovery of the affected region and the rebuilding of the
damaged residential, commercial, and public structures.15
Opponents of federal disaster insurance, however, say such measures conflict
with long-established sociological, economic, and actuarial principles that focus on
the “true”cost of government programs (the opportunity cost of the funds), the
foregone benefits of a competitive insurance marketplace (e.g., cost efficiency and
rate competition), and the absence of consumer choice (the ability to decide whether
to purchase coverage).16 Citing the development of innovative financial instruments
for natural disaster risk management and expanded reinsurance capacity, critics of
public insurance systems say there is no need for a federal insurance program at this
time. They insist that such programs shield the private sector from loss while
creating sizable taxpayer-financed subsidies that undermine private-sector incentives
for efficient risk management. Further, it has been argued that these programs
encourage population growth and development in high-risk, hurricane-prone areas
that should not be developed, and would allow insurers to “cherry pick” the best risks
and send the federal government the poor risks. Rather than providing insurance
protection for natural hazard losses, critics argue, the federal government should take
actions to expand private-sector capacity for insuring disaster losses.
Proponents of federal disaster insurance argued that such a scheme would
reduce dependence on “free” disaster assistance and support efficient risk
management by households and businesses.
The 2005 Hurricane Season
The 2005 hurricane season was the most destructive in recent U.S. history.
There were 27 named storms of which 14 were hurricanes. The 27 named storms
more than doubled the national average over the past five decades. According to
researchers at Tillinghast-Towers Perrin, a global actuarial, management, and
financial services consulting firm, over the past 50 years, there have been, on
average, 9.8 named storms, 5.8 hurricanes, and 2.3 intense hurricanes. Not all of
these storms made landfall. Some catastrophe risk modeling firms contend that while
the 2005 hurricane season was above normal, it was not so unusual. According to
their hurricane models, insurers should expect to see four hurricanes making landfall
in the United States approximately once every 12 years and this is within the range
to which most insurers manage their catastrophe risk.
15 Howard Kunreuther and Richard J. Roth, Sr., Paying the Price: The Status and Role of
Insurance Against Natural Disasters in the United States (Washington: Joseph Henry Press,
1998), p. 92
16 Kunreuther, p. 93.
CRS-9
Tables 3 shows that the three major hurricanes in 2005 (Katrina, Rita, and
Wilma) and three of four of the major hurricanes in 2004 (Charley, Ivan, and
Frances) accounted for six of the top 10 most costly catastrophes in U.S. history.
Before the 2005 hurricane season, Hurricane Andrew ranked as the single most costly
U.S. natural disaster at $22.9 billion in 2007 dollars. Insured losses from Hurricane
Katrina alone are estimated to be $43.6 billion. The three most destructive
hurricanes in 2005 together account for $60.5 billion. Total economic losses from
Katrina will likely exceed $200 billion. Despite the magnitude of the insured
property losses suffered in 2005, the insurance industry has the financial strength to
pay the claims from Hurricane Katrina.
Table 3. Ten Most Costly Catastrophes in the United States
($ billions)
Dollar Losses
When
In 2007
Rank
Date
Disaster
Occurred
Dollars
1
Aug. 2005
Hurricane Katrina
$41,100
$43,625
2
Aug. 1992
Hurricane Andrew
15,500
22,902
3
Sept.2001
WTC 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
Sep. 1989
Hurricane Hugo
4,195
7,013
9
Sep. 2005
Hurricane Rita
5,627
5,973
10
Sept. 2004
Hurricane Frances
4,595
5,043
Source: Insurance Services Office’s Property Claims Service; Insurance Information Institute.
Hurricane Katrina caused widespread damage to homes and businesses in six
states — Louisiana, Mississippi, Alabama, Florida, Tennessee, and Georgia. Much
of the damage from the storm was the result of flooding, rather than wind. In
response to the devastation from the 2005 hurricane season, rating agencies are
changing their methods of assessing the adequacy of an insurer’s capital.17 In the
past, rating agencies examined an insurer’s exposure to loss relative to a “100-year”
catastrophe event. That is, they looked at a disaster that occurred once every 100
years. The new capital adequacy threshold is 250-year events, because the 100-year
events are forecast to increase in frequency. Rating agencies are also considering the
potential losses from catastrophes in the aggregate and requiring an insurer’s estimate
of its probable maximum loss to include the increase in the cost of labor and
materials as the reconstruction cost rises.
Several insurance coverage issues have arisen in the wake of Hurricane Katrina.
For example, losses from floods are not covered under homeowners insurance
policies, and homeowners have filed lawsuits against insurers seeking to void the
17 Susanne Sclafana, “Big Cats Force Change in Disaster Models,” National Underwriter
Property and Casualty Edition, December 5, 2005, p. 20.
CRS-10
flood exclusion in homeowners policies. In other lawsuits, plaintiffs claim that
flooding was caused by negligence in the construction and maintenance of the levees
which broke and inundated the city rather than an “Act of God.” With respect to
business and commercial property losses, plaintiff lawyers representing homeowners
have pursued oil and chemical businesses, seeking compensation for damages. Many
of these businesses, which face longer and costly business interruption losses and
untold amounts of extra expenses incurred in an attempt to restore business
operations, are suing their insurers. Insurers are still assessing individual losses and
analyzing various scenarios that will affect ultimate claim payments.
The 2004 Hurricane Season
According to the National Oceanic and Atmospheric Administration (NOAA),
there were 12 named storms during the 2004 hurricane season, of which nine affected
the United States: three as tropical storms (Bonnie, Hermine and Matthew) and six
as hurricanes (Alex, Charley, Frances, Gaston, Ivan and Jeanne). Four of the
hurricanes (Charley, Ivan, Frances and Jeanne) made landfall as “major” or Category
3 or higher events on the Saffir-Simpson Hurricane Scale. Three other hurricanes
(Danielle, Karl, and Lisa) did not make landfall.18 The nine named storms that
affected the United States resulted in 21 Presidential declarations of major disaster
covering 12 states, Puerto Rico and the U.S. Virgin Islands.19 Florida was affected
the most by the four hurricanes followed by Alabama, Georgia, Pennsylvania, and
North Carolina.20
Table 3 shows that three of the four 2004 major hurricanes — Charley, Ivan,
and Frances — rank among the top ten for both the costliest U.S. hurricanes and
insured loss events in U.S. history. For the first time since 1886, three hurricanes —
Charley, Frances, and Jeanne — made landfall in the same state — Florida; Ivan
made landfall in Alabama, but continued its path across Florida.21 Meteorological
forecasters had correctly predicted above-normal activity during the 2004 hurricane
season,22 based on a trend of above-average activity during seven of the last nine
seasons. Insurers therefore had knowledge of and presumably were prepared for
18 The National Oceanic and Atmospheric Administration (NOAA) also reported that there
was subtropical storm Nicole and ten tropical depressions.
19 These 12 states are: Alabama, Delaware, Florida, Georgia, Louisiana, Mississippi, New
Jersey, New York, North Carolina, Pennsylvania, South Carolina, and Virginia.
20 For more information on the impact of the four hurricanes on Florida see Insurance
Services Office, Inc., Press Release, “AIR Analysis Concludes 2004 Hurricane Season Is
Not As Unusual,” available at [http://www.iso.com/press_releases/2004/11_03_04.html],
visited on January 31, 2008.
21 Matt Brady, “Insurers Post Record First-Half Profits,” National Underwriter: Property
and Casualty, October 25, 2004, p. 32.
22 Forecasters attribute the recent rise in Atlantic basin hurricane activity on long-term
climate patterns, including continuation of warmer-than-normal ocean temperatures across
the tropical Atlantic. The warmer waters are associated with circulation patterns that form
an above-average hurricane season.
CRS-11
these events. According to climatologists, the level of activities for 2004 was similar
to that of 2003, but consumers and insurers were spared huge losses in 2003 because
very few of the tropical storms and hurricanes made landfall in the United States.23
Thus, insurers faced limited losses from the 2003 hurricane season in terms of
damages relative to their activity, but that was not the case in 2004 when four major
hurricanes made landfall in August and September.
The Insurance Information Institute in New York indicated that the four major
hurricanes that struck Florida and other Gulf and Atlantic Coast states in 2004 caused
$21.0 billion in wind-related insured losses, and total claims filed of 2.2 million.24
Total economic losses were about $56 billion.25 The four major hurricanes, as a
whole, exceeded the property damages from the 9/11 terrorist attacks ($22.0 billion)
and Hurricane Andrew ($22.9 billion). In addition to insurance pay-outs, Congress
passed two emergency supplemental appropriations statutes that provided a total of
$17.535 billion to hurricane victims.26
Property and casualty insurance typically incur an underwriting loss on their
business and make up these losses on the investment of premiums and loss reserves.
Despite record catastrophe losses, the year 2004 was the first time the industry
managed an underwriting profit in 26 years. Policyholders surplus, a measure of
claims-paying capacity, increased to a record $521.8 billion in 2007 (3rd Quarter),
compared to $427.1 billion at year end 2006.
Unlike Hurricane Andrew that led to 11 insurer insolvencies and 63 insurers
announcing plans to withdraw from the Florida market or significantly curtailing of
new business, only one small insurer — American Superior Insurance Company —
became insolvent as a direct result of last year’s hurricanes. The 11 insolvencies
were the largest number of hurricane-related insolvencies in U.S. history.27 Several
insurers did have their financial strength rating downgraded by various rating
agencies, and at least four insurers have started canceling insurance policies in
23 “Dangerous Planet: Living on Borrowed Time,” Reaction, July 2004, p. 18.
24 Robert P. Hartwig, Catastrophes: Insurance Issues (Insurance Information Institute:
February 2005), available at [http://www.iii.org/media/hottopics/insurance/xxx], visited on
January 31, 2008.
25 Swiss Re, Natural Catastrophes and Man-Made Disasters in 2006: Low Insured Losses,
[http://swissre.com/resources/ce8f6a80455c6b9f8b2bbb80a45d76a0-sigma2_2007_e.pdf],
visited on January 31, 2008.
26 See CRS Report RL32581, Supplemental Appropriations for the 2004 Hurricanes and
Other Disasters, by Keith Bea and Ralph M. Chite.
27 The number of property/casualty insolvencies in any given year differ based on an
organization’s specific criteria for including a company in the insolvency count. The
National Association of Insurance Commissioners, for example, list an insurer as insolvent
when a company triggers some formal regulatory action in the calendar year because of
significant financial impairment. Other organizations like A.M. Best list an insurer as
insolvent when they meet the same criteria, but they count each company in a group.
CRS-12
Florida.28 In addition, several insurers have announced that they will no longer seek
new business in the state.
American Superior Insurance Company wrote homeowners insurance coverage
for nearly 60,000 Floridians and had a premium volume of $34 million, representing
less than 1 percent of total homeowners insurance premiums collected in Florida.29
The company voluntarily consented to be placed into rehabilitation by the Florida
Department of Financial Services, Division of Rehabilitation and Liquidation. Under
a plan of receivership, a state-appointed official takes over the company’s operations
and, in the event the insurer cannot be rehabilitated, could liquidate its assets to pay
policyholder’s claims. If the liquidated assets cannot pay all claims the Florida
Property and Casualty Insurance Guaranty Fund will pay the shortfall in claims up
to $300,000, and impose an assessment on all property insurers operating in Florida
to pay claims on behalf of the insolvent insurer. Insurers, in turn, can write off the
guaranty fund assessments against their state income taxes, thereby shifting some of
the cost of the insolvency to all taxpayers in the state.
Insurance Lessons Learned from
Hurricane Andrew (1992)
There is little doubt that property insured losses from the 2005 and 2004
hurricane seasons would have been even higher were it not for actions taken by
insurers, regulators, and state legislators to both protect the industry’s balance sheets
and stabilize the property insurance markets in the aftermath of Hurricane Andrew
in 1992.30 After Andrew, Florida faced a “capacity gap” — the difference between
the amount of capital (insurance) available and the demand for coverage — and a
sudden shortage of reinsurance for hurricanes. This situation meant major primary
insurers operating in coastal high-hazard Gulf and Atlantic Coast areas could not
adequately spread their catastrophe risks, which, in turn, forced many of them to stop
writing new policies in hurricane-exposed states or to shut down operations
altogether for fear of over-exposure, financial impairment, or even insolvency.
Insurers were caught off-guard by the $15.5 billion (1992 dollars) in insured
losses associated with Hurricane Andrew because of significant errors in actuarial
estimates of potential hurricane-related losses. Prior to Hurricane Hugo in 1989, the
insurance industry never suffered any loss over $1 billion from a single hurricane.
Further, most insurance industry experts estimated the probable maximum loss
(PML) for a single hurricane in the United States at between $8 and $10 billion, and
that such an event would occur only once in a century. Hurricane Andrew took
insurers and forecasters by total surprise. In hindsight, because of the lull in
28 Paige St. John, “Florida Insurers Start Pulling Out of State,” The Ft. Myers News-Press,
January 7, 2005, p. A1.
29 Theo Francis, “Hurricanes Claim Their First Victim in Insurance Field,” Wall Street
Journal, September 30, 2004, p. B2.
30 Theo Francis, “This Year’s Storms Fail to Blow Down Insurers,” The Wall Street Journal,
September 28, 2004, p. C3.
CRS-13
hurricane activity during the 1970s and 1980s, insurance policies were underpriced
and insurers accepted far more hurricane exposure than could be supported by their
capital resources (including reinsurance). Also, there were deficiencies in the storm-
resistant capabilities of homes in Florida as well as poor enforcement of building
codes in the region.
In response to post-Andrew insurance market disruption, state insurance
regulators undertook several steps to restrict insurers’ products, pricing, underwriting
decision and claims settlement practices for disaster coverage.31 In addition, the
South Florida Building Codes were extended statewide and the state legislature
established the Florida Commission on Hurricane Loss Projection Methodology to
review hurricane catastrophe models used for rate filings. These two major changes
were instrumental in defining how insurers process and analyze hurricane risk.
Insurers prospectively evaluated their catastrophe exposures in coastal areas for
the first time and discovered that the magnitude of risk was both unexpectedly high
and unacceptable, given the risk tolerances of management and the expected long-
term return on the business written in hurricane-prone areas in coastal states.32 The
concern was that insurers with excessive catastrophe exposures would have difficulty
achieving or maintaining profitability and balance-sheet strength, and this could lead
to rating downgrades, insurer insolvencies, and insurance availability problems.33
One major outcome of insurers’ assessment of catastrophe risk exposure was
that large national property insurers began forming single-state affiliate insurers to
protect the capital of the holding company. Also, with the approval of state
regulators, insurers began shifting the risk of windstorm losses away from
overexposed insurers to all property owners and other consumers (through
assessments from state-sponsored pools). This decision allowed consumers and
insurers to withstand hurricane-related losses in 2005 and 2004 with limited market
disruption in terms of policy cancellations, non-renewals and insurer insolvencies.34
31 For example, regulators sought to: (1) issue moratoriums disallowing cancellations and
non-renewals of homeowners insurance policies; (2) suppress homeowners insurance rates
in response to political pressure, but later approved rate hikes and special hurricane or
“wind” deductibles; and (3) open up the market to excess and surplus lines insurers and
state-sponsored insurance.
32 Rude Musulin, “Property Insurance Market Crisis,” Presentation before the Institute for
International Research, May 14, 1996, New York, NY.
33 Jeanne H. Dunleavy, Daniel, L. Ryan, and C. Brett Lawless, “Catastrophes: A Major
Paradigm Shift for P/C Insurers,” Best Week Property/Casualty Supplement: A Special
Report, March 25, 1996, p. 1.
34 On August 18, 2004, Florida implemented a moratorium prohibiting insurance companies
from non-renewing or cancelling the policies of homeowners hit by this year’s hurricanes.
Under the moratorium, residential insurers have been kept from dropping any policies, even
in cases involving nonpayment of premiums. Florida Treasurer Tom Gallagher announced
on November 16, 2004, that he wanted to extend the order beyond the end of November
because thousands of homeowners waiting on insurance checks would not be able to
complete repairs by that time and therefore not be able to get coverage elsewhere until the
(continued...)
CRS-14
The economic rationale for shifting the risk of windstorm loss to property
owners through state-sponsored insurance and reinsurance pools was that these pools
have a cost of capital advantage over private insurers. State-sponsored insurance
pools can offer coverage at a price below what the risk would normally require a
private insurer to charge. The pooling arrangement works because state insurance
pools can largely avoid the accounting and tax rules governing the private sector. A
state-sponsored insurance facility is able to defer part of the cost of capital to the
future by virtue of the government’s authority to issue public sector debt to pay
losses, and favorable tax treatment. But, as economists and financial analysts note,
there are limits to the ability of states to fund/capitalize insurance pools in advance
of catastrophe losses. That is, many consumers could face unpaid claims.
Insurance Market Response to Past Hurricanes
Insurers responded to Hurricane Andrew in 1992 by taking action in four areas:
hurricane insurance deductibles, a capital market for catastrophe securities, building
code regulation and construction standards, and catastrophe modeling and forecasting
tools. Collectively, these four marketplace changes allowed private insurers,
reinsurers, and state-sponsored insurance pools to withstand significant losses from
the 2004 hurricane season and to continue operating in disaster-prone states.
Hurricane Insurance Deductibles
Seventeen states and the District of Columbia now require property owners to
pay hurricane or windstorm deductibles from 1% to 15% of the insured value of the
property, depending on the type of home (e.g., mobile homes carry a higher
percentage deductible) and where the property is located, rather than traditional dollar
deductibles used for other types of claims, such as fire damage and theft.35
According to the Insurance Information Institute, the hurricane insurance deductibles
have had the beneficial effects of making insurance coverage more available in high-
risk areas, and getting customers more motivated to invest in disaster mitigation, such
as hurricane shutters, damage resistant windows, and homes fortified to withstand
severe storms.36 By imposing a higher deductible for windstorm-related losses,
property owners assume a greater share of the risks associated with living in high-risk
areas, and, therefore, they presumably take steps to mitigate potential losses.
34 (...continued)
repairs are finished.
35 These 17 states are: Alabama, Connecticut, Florida, Georgia, Hawaii, Louisiana, Maine,
Maryland, Massachusetts, Mississippi, New Jersey, New York, North Carolina, Rhode
Island, South Carolina, Texas, and Virginia.
36 For more information on windstorm deductibles see New York-based Insurance
Information Institute’s press release, dated September 20, 2004, “Insurance Deductibles
Apply for Each Claim,” available at [http://www.iii.org/media/updates/press.737890/],
visited on March 21, 2005.
CRS-15
The requirement that Florida homeowners pay a per-occurrence versus a
seasonal (aggregate) windstorm deductible emerged as an important state legislative
issue following the 2004 hurricane season.37 Some108,000 homeowners were struck
by two or more hurricanes, and approximately 36,000 policies had multiple
deductibles applied and the cost to policyholders of second and subsequent
deductibles may total about $70 million.38 Residential hurricane deductibles are
typically 2% of policy limits and may be as high as 5% of policy limits, or even
higher for certain policies.39 For this reason, the multiple deductible can result in
significant out-of-pocket expense for many policyholders.
After the 2004 hurricane season, the Florida Legislature passed legislation —
Hurricane Deductibles for Residential Insurance Policies (HB 9-A) — that
established a program to reimburse policyholders for financial hardships suffered due
to multiple hurricane deductibles being applied to their insured losses in 2004.
Under HB 9-A, policyholders of residential property insurance policies who paid two
deductibles in 2004 were eligible for reimbursement from the Department of
Financial Services up to $10,000 per storm, per policy, per structure, and up to
$20,000 if they paid three or more deductibles.
The Multiple Deductible Reimbursement Program is funded with $150 million
borrowed from the Florida Hurricane Catastrophe Fund (Cat Fund) to reimburse
residential property insurance policyholders .40 The borrowed funds would be repaid
over five years starting in 2006. The Cat Fund estimates that there was a statewide
average increase of 0.5% in homeowner rates to cover the payments.41
While insurers were generally pleased with the new law because they will not
have to reopen thousands of already settled claims, they had lobbied to make it easier
for insurers to be reimbursed by the Cat Fund. Currently, there is a $4.5 billion
threshold trigger before insurers can be reimbursed for losses under the Cat Fund’s
reinsurance agreement. Florida’s Chief Financial Officer, Tom Gallagher, had
recommended a plan to the Legislature’s Joint Select Committee on Hurricane
Insurance to reduce the Cat Fund retention to $4 billion for each of the two
hurricanes and $1 billion for the third and subsequent events in a season.42
37 Aaron DeSlatte, “Catastrophe Fund Stirs Debate,” The Florida Today, December 9, 2004,
p. 1.
38 For more information see, Office of the Florida State Senate Secretary, “Summary of
Legislation Passed: 2004-A Special Session,” available at [http://www.flsenate.gov], visited
on February 16, 2005.
39 It should be noted that $500 hurricane deductibles are still prevalent for homes and mobile
homes valued under $100,000.
40 For a summary of legislation passed by the Florida Legislature during the 2004-A Special
Session, see [http://flsenate.gov/Publications/2004A/Senate/reports/summaries/pdf/
sessum04A.pdf], visited on February 17, 2005.
41 Frank Matso Lysiak, “Bill Frees Catastrophe Fund to Reimburse Florida Deductibles,”
Best’s Review, January 2005, vol. 105, p. 10.
42 NAMIC Online, “Florida: Senators Hear Testimony on How Hurricane Affected the
(continued...)
CRS-16
Capital Market Instruments Linked to Catastrophe Risk
Insurers have traditionally used reinsurance to manage a portion of their
catastrophe risk. Insurers, reinsurers, and an increasing number of corporations came
to the realization beginning in the late-1980s that the traditional reinsurance
mechanisms were limited in their ability to provide coverage for catastrophic risk.43
Recognizing the limits of their ability to finance catastrophe risk, the high cost of
reinsurance, and the sheer size of the capital markets, insurers and investment banks
became more active in offering capital market instruments linked to catastrophe risk.
Investors are attracted to innovative financial instruments form natural disaster risk
management for several reasons, including their above-average risk-adjusted rate of
return versus the typical fixed income instruments and the fact that the rate of return
is not correlated with the returns associated with stock and bond portfolios.
The first risk-linked securities (called “catastrophe bonds”) were introduced in
1994, but it was not until 1997 that they gained some acceptance as catastrophe risk
financing alternatives. Total cat-bond issuance in 2005 is estimated at around $6
billion. Investors in these securities continue to demand a high-risk premium
because of their lack of familiarity with catastrophe risk and uncertainty about the
likelihood that these instruments will be triggered.44 The full acceptance of this new
asset class for securitization has been limited by: (1) the tax, cost and regulatory
treatment of the financial instruments — the so-called “special purpose reinsurance
vehicles” (SPRVs) — underlying the securitization; (2) the lack of standardization
in risk measurements; (3) lack of a generally-accepted index on which to base
payouts; and (4) high transaction costs relative to traditional reinsurance coverage.
Building Codes and Construction Standards
Disaster risk reduction requires effective enforcement of building codes, land-
use planning, environment risk and human vulnerability monitoring and safety
standards. In hurricane-prone coastal states like Florida, homeowners insurance rates
are now based on new building code standards and the structure’s ability to withstand
damage by high winds. In the 1980s, the insurance industry came to the realization
that the level of building code enforcement affected the cost of claims. It was not
until Hurricane Andrew in 1992, however, that a new organization, the Insurance
Institute for Property Loss Reduction (IIPLR), launched a study to develop better
wind and seismic building codes so structures could better withstand the force of
storms and earthquakes. The work of the IIPLR led to the development by Insurance
42 (...continued)
Industry, [http://www.namic.org/PrintPage.asp?ArticleID=7510], visited on January 31,
2008.
43 Paul R. Kleindorfer and Howard Kunreuther, “Challenges Facing the Insurance Industry
in Managing Catastrophe Risks”, In The Financing of Catastrophe Risk, ed., Kenneth A.
Front (Chicago, University of Chicago Press, 1999), p. 149.
44 Martin Grace, Robert W. Klein, and Richard D. Phillips, “An Economic Appraisal of
Securitizing Insurance Risk Via Onshore Special Purpose Vehicles,” Risk Management and
Insurance Review, 2002, vol 4, p. 33.
CRS-17
Service Office (ISO) of a building code compliance rating system. The ISO Building
Code Effectiveness Grading Schedule (BCEGS) assesses the building codes in effect
in a particular community and the community enforcement of these codes. The
BCEGS takes into account factors such as the size of the community’s building code
enforcement budget relative to the amount of building activity, the professional
qualifications of building inspectors, and past code enforcement levels. By
incorporating the BCEGS into the underwriting and pricing process, communities
now have the incentive to undertake mitigation activities such as requiring property
owners to use certain roofing material, the installation of hurricane shutters, and the
identification of appropriate load combinations for buildings.
With the availability of BCEGS, insurers and state insurance regulators
combined forces under the auspices of the National Association of Insurance
Commissioners (NAIC) to develop and encourage states to adopt model insurance
laws, regulations and guidelines that link insurance practices to building codes. The
Florida legislature requires insurers to reflect BCEGS in their rates. Insurers now
offer discounts on property insurance premiums to property owners and businesses
located in communities with enforced, up-to-date building codes that conform to
BCEGS standards. Communities with a BCEGS grade of 1 (reflecting exemplary
commitment to building-code enforcement), for example, can demonstrate better loss
experience, resulting in lower insurance premiums. Insurers may also impose
surcharges in communities where enforcement is lax. The BCEGS program was
initially implemented in states with high exposure to wind (hurricane) and seismic
exposure, but now is available throughout the rest of the country.
Catastrophe Modeling and Insurance Underwriting
Before Hurricane Andrew in 1992, most insurers had not used electronic
information processing systems to keep track of their potential hurricane loss
exposure and to help them make informed insurance underwriting decisions.45 After
the Andrew disaster there was a widespread use of catastrophe simulation modeling
— a type of modeling that allows insurers and regulators to better predict future
windstorm losses on the basis of current demographics and construction techniques,
rather than historical loss experience. Actuaries had gained access to sophisticated
statistical databases and computer modeling techniques that could integrate long-term
weather data, engineering studies of storm loss potential, and population trends.46 By
combining mathematical representations of the natural occurrence patterns and
characteristics of hurricanes, tornadoes, severe winter storms, earthquakes, and other
catastrophes, with information on property values, construction types, and occupancy
classes, these computer simulation models provide information concerning the
potential for large disaster losses before they occur.
45 Tom O’Brien, “Catastrophe Modeling for Corporate Risk Managers,” Risk Management
Magazine, May 2004, p. 18.
46 Michael Ha, “Catastrophe Modeling, Forecasting Tools More Sophisticated,” National
Underwriter: Property & Casualty/Risk & Benefits Management Edition, September 23,
2004, p. 17.
CRS-18
There are significant limitations to these types of computer modeling
techniques. For example, loss models work best when they are used to develop a
relative understanding of potential damage rates rather than absolute losses. Despite
the comparative wealth of data and knowledge about hurricanes and the
sophistication of insured loss models for these events, some experts believe that these
models are often wrong by an order of three, even if all the important event
characteristics are known. Thus, a model may predict that a given storm will produce
$300 million of insured losses, but the actual insured losses would vary from $100
million to $900 million.
Transferring Risk Through Insurance
Most existing structures in hurricane-prone areas are susceptible to hazard risks,
such as strong winds, storm surges, heavy rains, and flooding. Insurance as a risk
transfer mechanism can play a key role in helping to minimize disaster losses and
reduce the financial and economic impacts of disasters. The problem is that multiple-
peril insurance policies held by homeowners exclude damages caused by wind and
water damage. To fill this gap in coverage, state catastrophe funds, such as the
California Earthquake Authority and the Florida Hurricane Catastrophe Fund,
provide coverage for windstorm and earthquake hazards. In a similar fashion, flood-
related damages associated with hurricanes may be insured through a separate policy
offered by the federal National Flood Insurance Program (NFIP).
Hurricane Katrina demonstrated the importance of the NFIP for protecting many
families from financial ruin, but participation rates in flood-prone areas are under
30%. Many homeowners appear not to be insured despite a mandatory purchase
requirement as a condition of being eligible for a federally insured loan.
Federal Flood Insurance Program
Insurance against flood hazard is generally not available in the private insurance
market because flood risk is generally considered uninsurable: only people living in
flood zones could be expected to purchase flood insurance (adverse selection), and
these people would have frequent claims, thus making the coverage prohibitively
expensive.47 In addition, insurers generally lack the ability to spread risk sufficiently
to safeguard their assets against catastrophic flood losses. Therefore, as part of the
National Flood Insurance Act of 1968,48 Congress authorized the National Flood
Insurance Program (NFIP) to serve as an insurance alternative to disaster relief and
to meet the escalating costs of damage to buildings and their contents.
47 Some insurers provide coverage under homeowners insurance policies for backup of
sewers and drains. Coverage may also be provided for flood damage under the
comprehensive section of standard auto insurance policies and some coverage is available
under special commercial insurance policies.
48 P.L. 90-448; 83 Stat. 476.
CRS-19
Prior to 1968, the federal government responded to flooding on a national scale
through the building of flood control structures that restricted the flow of waters (e.g.,
dams, levees, and dikes) and providing disaster relief to flood victims. After decades
of federal expenditures for structural flood works and expanded disaster relief, the
focus shifted to flood insurance as a policy tool for reducing loss and for spreading
the risk of loss among individuals and businesses. It was expected that homeowners
and businesses would pre-fund their own losses by purchasing federal flood
insurance. At the same time, the program would encourage preventive and protective
measures to reduce future losses. A key mechanism for doing so was the
development of flood plain maps and the requirement that local communities restrict
development in areas most subject to flooding.
The NFIP provides subsidized, low-cost flood insurance to homeowners and
small businesses in flood-prone communities that have agreed to adopt and enforce
floodplain management and building code standards. Federal flood insurance is
available in each of the 50 States, the Virgin Islands, Puerto Rico, Guam, the District
of Columbia, and American Samoa to meet the escalating costs of repairing damage
caused by flood to buildings and contents.
The NFIP operates under a statutory mandate that premium charges for Pre-
FIRM risks — i.e., structures built before the issuance of a Flood Insurance Rate Map
(FIRM) or before 1975, whichever is later — must be reasonable. The subsidy is
provided by charging premium rates discounted from full risk rates. In order to make
up the premium shortfall from subsidizing premiums, the NFIP establishes a target
level of premium income for the program as a whole that accommodates the
combined effect of the portion of NFIP business paying less than full risk premiums
and the portion of the business paying full risk premiums.
Hurricane Katrina will ultimately cost the NFIP approximately $21 billion in
insured losses. In addition, both FEMA and the NFIP are under attack by property
owners and the insurance industry to improve the program to more appropriately
protect property values in flood-prone areas. Some members of Congress have called
for changes in the NFIP that include:
! expanding the mandatory flood zones from their current 100-year flood
zone level to the 500-year level;
! increasing the building and contents limits of flood coverage from
$350,000 for residential and $ 500,000 for commercial properties;
! adding business interruption insurance coverage.
Faced with the growing costs of federal expenditures on flood-related disaster
relief assistance, including insurance claim payments, and the cumulative impact of
low-intensity hurricanes on local economies (in terms of property damage and
subsequent reconstruction activity) Congress has continuously sought to strengthen
CRS-20
the operational and financial aspects of the NFIP.49 On June 30, 2004, President
Bush signed legislation to reauthorize the NFIP until September 2008 and to provide
states and local communities with an additional $40 million a year for mitigating
(i.e., buyouts, elevation or move the home) severe repetitive loss properties
(SRLPs).50
On September 27, 2007, the House approved H.R. 3121, the Flood Insurance
Reform and Modernization Act of 2007 to reform the program while retaining its
original intent to keep rates affordable for people to buy the insurance. H.R. 3121
would also increase the NFIP’s Treasury borrowing authority from $20.775 to $21.5
billion. On November 1, 2007, Senator Christopher J. Dodd introduced S. 2284 to
restore the financial solvency of the program and to forgive the debt.
In the absence of federal government intervention into the disaster insurance
market, several states — Florida, California, Hawaii, Louisiana — have had to
address the issue of “uninsurable risks,” meaning risk that cannot get coverage from
private insurers in the “voluntary market.” States with a high risk of natural disasters
have created catastrophe funds or residual markets to deal with the unavailability and
unaffordability of property insurance. The residual market initiatives take on various
forms, such as: (1) Fair Access to Insurance Requirement (FAIR) Plans that are used
to cover “hard to insure” exposures; (2) Beach & Windstorm Plans that operate by
spreading the risks among insurers operating in the state; (3) Marketing Assistance
Plans (MAP) that address short-term insurance availability and affordability
problems in a state; and (4) provision for the operation of surplus lines. Both the
property insurance residual markets and catastrophe funds as state-sponsored loss-
sharing mechanisms will be discussed in the next section.
State-Sponsored Lost-Sharing Mechanisms
In states where insurers in the private market have reached the limits of their
willingness or ability to provide coverage for homes and businesses in high risk
areas, the state has created catastrophe funds and property residual insurance markets
( i.e., Fair Plan and Beach & Windstorm Plans), marketing assistance plans, and
provisions for surplus line operations that serve to stabilize the property insurance
market — without the involvement of the federal government. Following is a brief
discussion of these state residual insurance markets.
Fair Plans. The District of Columbia and 34 states have Fair Access to
Insurance Requirement (FAIR) plans that make property insurance available to
applicants on eligible property located in coastal areas who have been unable to
secure such insurance in the normal insurance market. FAIR Plans are syndicated
associations of property insurers doing business under the auspices of the state
insurance regulator. Although the FAIR Plans act as a single insurer, participating
companies actually share on a pro rata basis all of the premiums as well as the profits
or losses and expenses incurred.
49 Robert T. Burris et al, “Impact of Low-Intensity Hurricanes on Regional Economic
Activity,” Natural Hazards Review, August 2002, p. 118.
50 P.L. 108-264; 118 Stat. 712.
CRS-21
The concept for FAIR Plans emerged in response to urban riots and civil
disorder in the 1960s and the withdrawal of insurers from the property insurance
market in communities with a high potential for loss. Congress enacted the Housing
and Urban Development Act of 196851 which sought to ensure the availability and
affordability of fire, crime, and other property insurance in high-risk urban areas by
offering federal riot reinsurance to property insurance companies operating in states
that voluntarily adopted a FAIR Plan.52
Beach and Windstorm Insurance Plans. In 1969, following Hurricane
Camille, the first Coastal or “Beach” Pool was created to address the shortage of
windstorm insurance in areas vulnerable to hurricane losses. Today, nine states have
formed Beach and Windstorm Plans which provide coverage for the wind peril alone
in designated risk-prone coastal areas.53 In some states the FAIR Plan serves as the
windstorm plan. Under a state-sponsored windstorm pool, the wind coverage is
isolated, and a separate policy is issued for this peril by the private insurer.
Windstorm pools typically purchase reinsurance to cover future losses.
Florida’s Citizens Property Insurance Corporation. On July 1, 2002,
the Florida Legislature passed a law that created the Citizens Property Insurance
Corporation (Citizens) as a market of last resort for residential and commercial
residential54 coverages in high-risk areas where the property owner is unable to
procure insurance in the open, private insurance market.55 Citizens was created with
the merger of the two existing property residual markets: Florida Residential Property
and Casualty Joint Underwriting Association (FRPCJUA) and the Florida Windstorm
Underwriting Association (FWUA).
Citizens operates under the authority of a seven-member Board of Governors,
approved by the State Treasurer. The State Treasurer also appoints a technical
advisory Board of Governors to provide information and advice to the seven-member
Board of Governors. All revenues, assets, liabilities, losses, and expenses of Citizens
are divided into three separate accounts: (1) a personal lines account for personal
residential polices issued by Citizens or the FRPCJUA, and renewed by Citizens ,that
provides comprehensive, multi-peril coverage on risks which are not located in areas
eligible for coverage in the FWUA (and for such policies that do not provide
coverage for the peril of wind); (2) a commercial lines account for commercial
51 P.L. 90-448; 83Stat. 476.
52 The Federal Riot Reinsurance Program was terminated on September 30, 1984, due to the
small number of insurers buying the reinsurance.
53 These nine states are: Alabama, Georgia, Florida, Hawaii, Louisiana, Mississippi, North
Carolina, South Carolina, and Texas.
54 Commercial residential simply refers to small business located in areas primarily zoned
residential.
55 Residential coverage includes both personal lines residential coverage (which consists of
the type of coverage provided by homeowner’s, mobile home owner’s, dwelling, tenant’s,
condominium unit owner’s, and similar policies) and commercial lines residential coverage
(which consists of the type of coverage provided by condominium association, apartment
building, and similar policies).
CRS-22
residential policies; and (3) a high-risk account for personal residential policies and
commercial residential and commercial non-residential property policies.
Citizens is authorized by statute to issue bonds and impose emergency
assessments on all licensed property insurers in the state. For the most part, these
assessments are eventually passed onto consumers. In order to maximize the
financial resources to pay claims following a catastrophic hurricane, Citizen’s income
and the interest on the debt obligations issued by the corporation are exempt from
federal income taxation.
Although the presence of Citizens in the market has stabilized the availability
and pricing of insurance in coastal areas of Florida, legislators and regulators have
concerns about the growth in the number of policies issued during the past few years
as well as the fact that all insurance customers statewide are responsible for a
significant deficit.56
On January 22, 2007, the Florida Senate and House passed a final bill which
goes to the Governor for signature. The Governor has not indicated whether he will
sign the bill. In general, the Florida legislature has (1) expanded Florida Citizens
exposure and decreased rates, (2) expanded the Florida Hurricane Catastrophe Fund,
(3) expanded the assessment authority of FIGA, and (4) rolled the Commercial Lines
JUA into Citizens.
Florida Hurricane Catastrophe Fund. In 1993, the Florida Legislature
created the Florida Hurricane Catastrophe Fund — “Cat Fund” — in response to
insurers’ concerns about actual and threatened catastrophic losses to property in the
state from hurricanes, and their unwillingness or inability to provide property
insurance coverage to the extent sought and needed. The Cat Fund was therefore
established as a tax-exempt source of reimbursement to property insurers for a
selected percentage of hurricane losses above the insurer’s retention (deductible).
The reinsurance provided by the Cat Fund is designed to stabilize the residential
property insurance market in the event of a major hurricane by offering relatively
inexpensive reinsurance to property insurers and the state’s insurers of last resort —
Citizens. The cost of this reinsurance is below what the private reinsurance market
charges because the Cat Fund is not only exempt from federal income tax, but also
the state’s income tax and premium tax. The Cat Fund was the first program in the
United States in which a state provided for tax-exempt accumulation of private cash
to pay for major disasters. The relatively inexpensive reinsurance sold by the Cat
Fund to Citizens (and other residential property insurers) allows them to write more
residential property insurance in the state — than otherwise would be the case — and
also acts to lower premiums for consumers.
On June 1, 2004, the Florida Legislature expanded the overall claims paying
capacity of the Cat Fund from $11 billion to $15 billion. The $15 billion amount is
financed primarily through reinsurance premiums paid by primary insurers (based on
56 David Sedore, “Citizen may Bill $60 for Deficit,” Palm Beach Post, February 16, 2005,
p. C1.
CRS-23
their exposure to hurricane losses) and investment income.57 Retained earnings are
held in a reserve fund account that accumulates, along with investment earnings, on
a tax-free basis. In the event cash reserves are insufficient to pay claims, the Cat
Fund could issue state government revenue bonds or other debt instruments to raise
billion of dollars for claims payout.
Under the Cat Fund’s reinsurance arrangement, private insurers and Citizens
are reimbursed for as much as 90 percent of insured hurricane losses in excess of a
$4.5 billion per storm deductible, up to a total of $15 billion each year. Insurers can
choose from three reimbursement coverage options — 45 percent of losses over the
retention, 75 percent or 90 percent — depending on their risk tolerance levels and
how much they want to pay for reinsurance. The $4.5 billion is an industry
deductible. Each insurer has an individual deductible, which is its proportionate
share of the $4.5 billion industry aggregate. This individual insurer deductible allows
smaller insurers that suffer unusually heavy losses to qualify for reimbursement,
while the industry overall might not. Insurers also have an individual maximum
coverage which is their individual share of the $15 billion maximum industry
aggregate collected by the Cat Fund. Insured losses above $15 billion would be
covered by the insurer’s high layer reinsurance and their surplus or reserves.
Louisiana Citizens Property Insurance Corporation. On January 1,
2004, the Louisiana legislature merged the Louisiana Joint Reinsurance Plan (FAIR
Plan) and the Louisiana Insurance Underwriting Plan (Beach and Windstorm Plan)
to create the Louisiana Citizens Property Insurance Corporation as an insurer of last
resort for property owners unable to obtain insurance in the state. Policies in force
at the time of the merger were to be handled by their respective plans. New
insurance business is being placed with Louisiana’s Citizen. This new entity can
build up reserve funds on a tax-free basis to pay claims after a natural disaster. In the
event the fund falls short, the state can issue revenue bonds to pay claims. Private
insurers are responsible for retiring the bonds, but can pass on the costs to
policyholders in the form of a surcharge.
Hawaii Hurricane Relief Fund. After Hurricane Iniki struck in 1992, the
Hawaii legislature created a Hawaii Hurricane Relief Fund (HHRF) to provide
windstorm coverage for residential properties in Hawaii. Under the state-sponsored
insurance scheme, insurers are allowed to sell homeowners insurance with a
hurricane exclusion. Each participating insurer in the state then acts as a servicing
insurer for the HHRF, issuing the insured a separate hurricane policy and collecting
a separate premium that is then forwarded to the HHRF.
57 This debt may be serviced with future reinsurance premiums collected by the Cat Fund
and/or post-hurricane assessments levied on all property and casualty insurers and, hence
reflected in future property and casualty insurance premiums. In the event that premiums
and proceeds that can be raised through the issuance of tax-exempt revenue bonds are
insufficient to address a catastrophic loss, claims submitted by insurers to the Cat Fund are
paid on a pro-rata basis. Because the annual post-hurricane assessments are subject to a cap,
losses generated by a major catastrophe could be paid over a number of years following the
event. The Cat Fund is never obligated to pay more than its assets and borrowing capacity
permit. The state is not liable for unpaid claims.
CRS-24
The HHRF receives ongoing revenue from hurricane premiums, and insurance
companies post-hurricane assessments on property business and mortgage recording
fees. The plan provides coverage for losses up to approximately $2 billion in
residential damages from hurricanes. Homeowners are responsible for the first 10%
in losses from a major hurricane. Private insurers participating in the HHRF would
be responsible for the next loss layer after the insured’s deductible. The Fund uses
some of the hurricane premium revenues to buy reinsurance, which covers the third
loss layer. The last layer of coverage comes from lines of credit, which are secured
by future surcharges on all property and casualty premiums. If losses exceed the total
coverage amounts, claims are paid on a pro-rata basis.
Market Assistance Plans (MAP). Four states — Florida, New York, New
Jersey, and Texas — have established Marketing Assistance Plans (MAPs) as service
organizations designed to assist consumers in obtaining property and casualty
insurance coverage from authorized insurers in the private market. These
organizations are voluntary mechanisms coordinated by private insures and agent
groups in cooperation with the state insurance regulators to provide insurance when
there is a “temporary” market failure. The MAPS collect and maintain information
on agents and insurers writing certain coverages.
MAPS are typically administered by insurance agents’ associations that will
assign insurance applicants who are declined coverage in the voluntary market to
participating insurers that agree to accept applicants on a proportional market-share
basis. Some states have created “Property Protection Programs” in conjunction with
MAPs to provide insurers with financial incentives (e.g., state premium tax credits)
to underwrite basic residential insurance coverages which can be tailored to fit the
specific needs of residents in a particular community exposed to catastrophe risk.
Surplus Lines Insurance
Any risk for which insurance is not available through a company licensed in the
applicant’s state (an “admitted” insurer) may be covered by a surplus lines insurer.
The business is placed with a “non-admitted” insurer in accordance with surplus or
excess lines provision of state insurance laws. Regulators allow catastrophe
insurance or “hard-to-place” coverage such as insurance for antique cars to be sold
in a less regulated environment on a surplus lines basis because of the unusual nature
of the risks, and the need for greater flexibility in policy terms and pricing. Also,
coverage may be offered on a surplus lines basis if the voluntary or involuntary
markets will not write expensive homes in a high-risk community.
Challenges for the 110th Congress and Beyond
In 2005, despite the ability to better predict and manage individual insurance-
company hurricane exposures, the property insurance industry, as a whole, still faces
the long-term challenge of maintaining viable insurance markets following a mega-
catastrophe that could threaten the solvency and claims-paying ability of the
insurance industry. Complicating this challenge is the recognition that the country
is increasingly exposed to greater hurricane-related losses as a result of three major
CRS-25
ongoing developments: (1) rapid expansion of the population into areas that are
susceptible to natural disasters; (2) rising property values in coastal areas; and (3)
climatological and environmental changes.
Population Growth and Coastal Development
With a significant percentage of our population now living in hurricane-exposed
areas, even larger insured property losses from hurricanes are possible.58 The
American population is migrating toward the coasts at a rapid rate, placing people
and property investments at risk of loss. In the last three decades, the nation’s
shorelines have come under increasing pressure from population growth and
development, and this has profound consequences for the insurance industry (through
higher losses) and the federal budget (through emergency supplemental
appropriations). As an illustration, according to the 2000 U.S. Census data, 55% of
the U.S. population live within 50 miles of a coastline (including Great Lakes
shorelines). The National Oceanic and Atmospheric Administration (NOAA) reports
that the population density per square mile in hurricane-prone Southeast coastal areas
increased 129%, versus 38% in the total U.S., during the 30-year period from 1960
to 1990.59 The Insurance Services Office (ISO) found that from 1970 to 1990, the
Southeast Atlantic Coast had a nearly 75% increase in population density, far
surpassing the countrywide increase of more than 20%.60
Given this trend in population growth and coastal development, policymakers
have become increasingly aware of the erosion risks facing homeowners and
communities due to high intensity storms and coastal flooding.61 It is not uncommon,
for example, for a hurricane or severe coastal storm to cause the coast to erode 100
feet or more in a single day. This situation has led to debate over the economic
consequences of erosion and the use of federal programs, such as the National Flood
Insurance Program (NFIP), to address the coastal erosion problem. The principal
concern is that while the NFIP covers erosion damage that occurs in connection with
floods, it does not account for erosion in setting flood insurance rates in coastal areas.
The insurance industry and FEMA have both taken steps to address coastal
erosion. Insurers responded to coastal erosion risk by making policyholders
vulnerable to windstorms pay more of the cost of living in hurricane prone areas. For
example, insurers now impose hurricane deductibles equal to a percentage of the
structure’s insured value and establish rates for windstorm coverage based on the
58 Rude T. Musulin, “Would a Federal Role in Disaster Protection Be a Catastrophe?”
Contingencies, November 29, 2003, p. 28.
59 Conning & Company, Lighting Candles in the Wind: Industry Response to the
Catastrophe Problem. Hartford, Connecticut. November 1994, p. 27.
60 Ibid.
61 The National Flood Insurance Reform Act of 1994 required that FEMA submit a report
evaluating the economic impact of erosion on coastal communities and the NFIP. The
study, which was written by the Heinz Center, recommended that FEMA develop maps that
identify coastal erosion hazard areas and include the cost of expected erosion losses when
setting flood insurance rates for coastal areas.
CRS-26
structure’s ability to withstand damage from high wind. FEMA has begun to prepare
and disseminate maps showing areas subject to erosion, created and imposed a
mandatory surcharge for erosion on flood insurance in Coastal High Hazard Zones,
and, with the passage of the Flood Insurance Reform Act of 2004, provided
relocation assistance and/or buyouts.
The National Coastal Zone Management Program (CZMP) is a federal-state
partnership authorized by the Coastal Zone Management Act62 to encourage coastal
states to develop and implement coastal zone management plans. Some 34 states and
territories participate in the CZMP.63 The CZMP is designed to encourage the states
to work with the federal government in finding a balance between protecting the
coast and preserving the human uses that depend on the environment. The CZMP
supports states through financial assistance and technical services and information.
In addition, the NFIP is managed in a manner that is consistent with the criteria and
standards established for the federally approved state coastal zone management plans.
Rising Property Values in Coastal Areas
Along with rising coastal population growth, there has been a tendency for
coastal development to consist of relatively more expensive properties.64 Table 1
shows that $6.9 trillion of the estimated $19 trillion of insured coastal properties are
vulnerable to hurricane risk. Whether provided by an insurer’s surplus, reinsurance
agreements, or securitized insurance instruments, capital is needed to underwrite
property insurance covering potential losses from weather and climate events. It
became obvious after Hurricane Andrew that exposures in disaster-prone areas were
far beyond the capital that was available from existing sources before the event.
Insurers have responded to the urgency of strengthening their capital position
with reforms of insurance systems. They have also played an active role in
encouraging the development of better wind and seismic building codes so future
construction could better withstand the force of hurricanes and earthquakes. It takes
time, however, to implement and realize the results of construction standards
nationwide. Because most of the building in coastal areas were constructed in the
1970 through 1990 period, when these building standards did not apply, most homes
remain vulnerable to damages from natural disasters. In addition, insurers have
shifted more responsibility for catastrophe damages to the property owners. They
have accomplished this by requiring higher deductibles and employing computer-
generated rates — rates which are based on a simulation of various scenarios
involving a structure’s ability to withstand damage by high winds and from water
damage.
62 P.L 92-583; 86 Stat. 1280.
63 For more information on Coastal Zone Management, see National Ocean Administration
Agency’s Office of Ocean and Coastal Resources Management, “Celebrating 30 Years of
Coastal Zone Management Act,” available at [http://www.ocrm.nos.noaa.gov/czm], visited
on March 21, 2005.
64 Musulin, p. 30.
CRS-27
Climatological and Environmental Changes
There is a growing body of scientific evidence suggesting that the global climate
may be changing (i.e., global warming), with more frequent extreme weather events
occurring, thereby increasing the incidence and severity of natural disasters such as
hurricanes and floods.65 According to Dean John Dutton of Pennsylvania State
University and the National Oceanic Administration Agency (NOAA), up to $2.2
trillion of the U.S. economy are believed to be affected annually by weather and
climate events. At the core of the global warming debate is the belief that human-
derived so-called “greenhouse gases” emissions have risen in recent decades,
resulting in a dramatic rise in both the temperature at the earth’s surface and the
frequency and severity of hurricanes, windstorms and floods.
The basic assumptions underpinning the pricing of insurance against hurricanes
does not take into account changes in the global climate. Moreover, insurers
traditionally assume that the average insured losses over a recent historical period
accurately reflect future losses over some arbitrary future period. This assumption
works well with automobile insurance and other widely distributed, independent risk,
but is less appropriate for low-frequency/high consequence events like hurricanes.
In other words, the average activity in any arbitrary period of the past is not
necessarily a good predictor of future activity. The inability to predict future
disasters will impact the capacity for disaster financing as part of a comprehensive
disaster management approach.
Issues and Policy Options
The devastation caused by Gulf Coast hurricanes in 2005 has stimulated an
important public dialogue about the efficacy of our current system for managing and
financing natural catastrophe risk. A central challenge facing members of the 109th
Congress is determining what public and private initiatives are needed to mitigate
and finance losses stemming from future mega-catastrophes. Is there a need to
improve the nation’s ability to finance catastrophe risk and, if so, how? More
specifically, Congress might be called upon to determine the appropriate roles and
policies of the public and private sectors in addressing disaster risks, how they affect
hurricane risk, and how they might be restructured to better achieve social objectives.
Three points of view usually emerge when debating the catastrophe funding
problem. One view is that catastrophes (e.g., hurricanes) are “uninsurable” in the
private sector and the federal government should directly take over underwriting
insurance. This view is not widely supported because the private sector has access
to capital market resources that can be used to fund the cost of a catastrophic
hurricane.
A second point of view argues that catastrophe risk is fundamentally uninsurable
and federal involvement in the catastrophe insurance market is needed before a really
65 Kelly Quirke, “Global Warming and Increasing Catastrophe Losses,” Journal of
Insurance Regulation, Summer 1994, p. 451.
CRS-28
“big” event occurs only at the higher layers of coverage.66 A consensus among
insurance and public policy experts seems to have emerged that a public-private
partnership in financing a mega-catastrophe over the $60-100 billion threshold might
eventually be needed. As demonstrated by the 2005 and 2004 hurricane seasons,
insurers and reinsurers are able to handle a series of catastrophes with insured
damages as high as $60 billion. Some insurance market analysts believe the industry
can even handle a single insured event approaching $100 billion in claims.
A public-private catastrophe insurance partnership would leave the private
sector responsible for underwriting property insurance, and the federal government
responsible for providing capital only where consumers and the private insurance and
capital markets are unable to do so. The government could facilitate more effective
risk-spreading, which can be achieved by more effective pooling of losses over time
and broader pooling of losses among risks. A more effective pooling of losses over
time could be achieved by borrowing mechanisms and tax deferral of loss reserves
for natural disasters. Broader pooling of losses among risks could be facilitated by
requiring property owners to purchase insurance against natural hazards. Economists
note, however, that any scheme that imposes a mandatory insurance requirement for
all property owners could create economic distortions, such as cross subsidization of
risks, where low risk individuals subsidize high risk individuals.
A third point of view presupposes that the financial resources available in the
private sector are sufficient to make federal involvement unnecessary at this time.
There have been many proposals considered by Congress during the last three
decades to address various types of catastrophic losses.67 Members of the 109th
Congress could pursue any one or all of the following policy options if it could be
shown that potential losses from hurricane-related hazards are beyond the capacity
of private markets to diversify such risks:
! establish an emergency reserve fund to provide timely financial
assistance in response to domestic disasters and emergencies;
! provide financial backstop or guarantees to innovative financial
instruments and activities involving insurance interstate compacts
between states set up to address regional exposure to catastrophic
losses involving hurricanes in the Gulf and Atlantic coast states, and
earthquakes in the Pacific northwest states and other areas exposed
to similar hazard risks;
! develop a large regional pool for hurricane insurance with potential
benefits of spreading the risks across impacted states. Currently,
66 For more information on this argument, see; James E. Rutrough, “Funding Major
Disasters with Traditional Insurance,” in Financial Risk Management for Natural
Catastrophes: Proceedings of a Conference Sponsored by Aon Group Australia Limited,
Neil R. Briton and John Oliver, ed., (Brisbane, Australia: Griffith Uni Print, 2002), p. 1
67 For more information, see CRS Report RL33086, Hurricane Katrina: Insurance Losses
and National Capacities for Financing Disaster Risk, by Rawle King.
CRS-29
each state relies on its own financial resources to develop a
substantial pool of funds for insurance. Some supporters of this
policy option have even suggested pooling risks with the same
characteristics as hurricanes, such as earthquakes, volcanos, and
tsunamis, into a single national hazard insurance program designed
to solve the catastrophe funding problem;
! establish a federal hurricane program to provide reinsurance to state-
sponsored insurance programs;
! establish an explicit federal windstorm insurance program, similar
to the National Flood Insurance Program;
! focus on tax policy to allow insurers to create tax-deferred reserves
to fund future catastrophe losses from natural disasters;
! encourage innovative new financing mechanisms of insurance,
reinsurance and capital markets to mitigate and diversify disaster
risk; and
! implement a comprehensive national disaster mitigation policy
strategy for reducing future losses.
These approaches have been debated in previous Congresses and some have
enjoyed bipartisan support; however, no consensus emerged, largely because of
concerns that such approaches would (1) encourage home construction in high-risk
areas, (2) serve as a tax giveaway to rich insurers, (3) expose the federal treasury to
large contingent costs at a time of budgetary deficits, or (4) give a competitive
advantage to certain segments of the insurance industry.
CRS-30
Conclusion
During the 2005 hurricane season, Hurricanes Katrina, Rita, and Wilma
demonstrated both the destructive nature of natural disasters and the importance and
shortcomings of insurance as a major financial source for post-disaster economic
recovery. As demonstrated by the 2005 and 2004 hurricane seasons, the industry can
readily handle a series of events with insured damages as high as $60 billion.
Insurers are expected to easily cover the $66.1 billion in insured losses from the 2005
catastrophes, including Hurricane Katrina, without threatening the industry’s
solvency or claims-paying ability. Many questions, however, have been raised about
the industry’s continued willingness and ability to sell property insurance coverage
in hurricane-prone states. While some insurance market experts believe insurers and
reinsurers can even handle a single insured event approaching $100 billion, this
amount of protection might not be enough for the potential mega-catastrophe. For
example, catastrophe modeling firms note that a repeat of the 1938 Category 3
hurricane that hit the Northeast could cause over $300 billion in possible damages.
As development increases in coastal areas, a mega-catastrophic hurricane could result
in huge government outlays for disaster assistance and present insurers with
significant financial hazards, such as the risk of insolvency, a rapid reduction of
earnings and statutory surplus, forced asset liquidation to meet cash needs, and
ratings downgrade.
To the extent property insurance markets fail to offer adequate levels of
coverage after a catastrophic hurricane, and the federal government avoids the
disaster insurance market (with the exception of terrorism risk insurance), the states
will likely continue to offer various, loss-sharing mechanisms that provide
catastrophe insurance or reinsurance coverage at subsidized rates. In a similar
manner, the federal government will continue to offer flood insurance under the
National Flood Insurance Program (NFIP) to offset repair and rebuilding costs in
flood-prone regions.
Members of the 110th Congress will likely be called upon to determine whether
there is a need to improve the nation’s ability to finance catastrophe risk and, if so,
how. Previous Congresses responded to similar concerns by considering legislation
to create a federal catastrophe reinsurance program for residential property. But,
despite broad support for several bills over the past few Congresses, the full Congress
did not authorize a federal reinsurance program until the enactment of the Terrorism
Risk Insurance Act of 2002.
Finally, most observers would agree that for the very highest layers of
catastrophe risk, the government (and consequently the taxpayer) is now, by default,
the insurer of last resort. In the 110th Congress, any one of a number of policy
options could be pursued, but passage will likely be based on whether it can be
shown that potential losses from hurricane hazards are beyond the capacity of private
markets to diversify catastrophe risks. Members will likely be grappling with several
CRS-31
policy questions. For example, will reinsurance and securitization be enough to
maintain insurance solvency after a catastrophic hurricane? How can the various funding
sources available for catastrophe insurance be expanded and refined to cope with a
catastrophic hurricane? And lastly, what role, if any, should the federal government play
in catastrophe insurance