This is the accessible text file for GAO report number GAO-08-7 entitled 'Natural Disasters: Public Policy Options for Changing the Federal Role in Natural Catastrophe Insurance' which was released on December 26, 2007. This text file was formatted by the U.S. Government Accountability Office (GAO) to be accessible to users with visual impairments, as part of a longer term project to improve GAO products' accessibility. Every attempt has been made to maintain the structural and data integrity of the original printed product. Accessibility features, such as text descriptions of tables, consecutively numbered footnotes placed at the end of the file, and the text of agency comment letters, are provided but may not exactly duplicate the presentation or format of the printed version. The portable document format (PDF) file is an exact electronic replica of the printed version. We welcome your feedback. Please E-mail your comments regarding the contents or accessibility features of this document to Webmaster@gao.gov. This is a work of the U.S. government and is not subject to copyright protection in the United States. It may be reproduced and distributed in its entirety without further permission from GAO. Because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Report to the Ranking Member, Committee on Financial Services, House of Representatives: November 2007: Natural Disasters: Public Policy Options for Changing the Federal Role in Natural Catastrophe Insurance: GAO-08-7: GAO Highlights: Highlights of GAO-08-07, a report to the Ranking Member, Committee on Financial Services, House of Representatives. Why GAO Did This Study: In recent years, much attention has been focused on the roles that the private sector and federal government play in providing insurance and financial aid before and after catastrophic events. In this context, GAO examined (1) the rationale for and resources of federal and state programs that provide natural catastrophe insurance; (2) the extent to which Americans living in catastrophe-prone areas of the United States are uninsured and underinsured, and the types and amounts of federal payments to such individuals since the 2005 hurricanes; and (3) public policy options for revising the federal role in natural catastrophe insurance markets. To address these questions, GAO analyzed state and federal programs, examined studies of uninsured and underinsured homeowners and federal payments to them, identified and analyzed policy options, and interviewed officials from private and public sectors in both high- and low-risk areas of the United States. GAO also developed a four-goal framework to help analyze the available options. What GAO Found: The federal government and some states have developed natural catastrophe insurance programs that supplement or substitute for private natural catastrophe insurance. These programs were created because homeowner coverage for catastrophic events is often not available from private insurers at prices deemed affordable by insurance regulators. Large losses associated with natural catastrophes are some of the biggest exposures that insurers face. Particularly in catastrophe-prone locations, government insurance programs have tended not to charge premiums that reflect the actual risks that homeowners face, resulting in financial deficits. After a resource-depleting disaster, the programs have postfunded themselves through, among other sources, payments from insurance companies and policyholders and appropriations from state and federal taxpayers. Large numbers of Americans are not insured for natural catastrophes. Homeowners may not purchase natural catastrophe insurance because doing so is voluntary and they may not believe that the risk justifies the expenditure. In addition, some homes may be underinsured—that is, not insured for the full replacement value. GAO estimates that the federal government made about $26 billion available to homeowners who lacked adequate insurance in response to the 2005 Hurricanes Katrina, Rita, and Wilma. Given the unsustainable fiscal path of federal and state governments, they will be challenged to maintain their current fiscal role. As Congress reevaluates the role of the federal government in insuring for natural catastrophes, Congress is faced with balancing the often- competing goals of ensuring that citizens are protected and limiting taxpayer exposure. This report examines seven public policy options for changing the federal government’s role, including establishing an all- perils homeowner insurance policy, providing reinsurance for state catastrophe funds, and creating a mechanism to provide federal loans for state catastrophe funds. Each option has advantages and disadvantages, especially when weighed against competing public policy goals. For example, establishing an all-perils homeowner policy is a private sector approach that could help create broad participation. But low-income residents living in parts of the United States with high catastrophe risk could require subsidies, resulting in costs to the government. Similarly, federal reinsurance for state programs could lead to broader coverage, but could displace private reinsurance. GAO also identified several policy options for tax-based incentives for insurance companies, homeowners, investors, and state governments. But these options, which could help recipients better address catastrophe risk, could also result in ongoing costs to taxpayers. While some options would address the public policy goals of charging risk-based rates, encourage broad participation, or promote greater private sector participation, these policy goals need to be balanced with the desire to make rates affordable. Selected Advantages and Disadvantages of Options for Changing the Federal Role in Natural Catastrophe Insurance: Option 1: All-Perils Homeowners Insurance Policy; This option would create a homeowner insurance policy that would provide coverage against all types of natural catastrophes. Advantages: * A mandatory all-perils policy could eliminate the problems of uninsured property owners and adverse selection; * A mandatory all-perils policy would end homeowners’ uncertainty about coverage for some perils. Disadvantages: * The all-perils option could require government subsidies for low- income property owners; * Premiums for an all-perils policy could be more expensive than current homeowner policy premiums, and these premium increases could be seen as unfair. Option 2: Federal Reinsurance for State Catastrophe Funds; This option would create federally backed reinsurance policies for state catastrophe funds. In one version of this option, states would create catastrophe funds that would be reinsured by the federal government. In another version, the Secretary of the Treasury would create an auction process for the sale of reinsurance contracts to private and state insurers and reinsurers. Advantages: * The federal reinsurance option could lead to greater participation from private insurers; * This option would not use tax dollars if premiums were risk-based. Disadvantages: * Federal reinsurance could compete with the private reinsurance sector; * Federal reinsurance could create inequities among states because of geographical differences in natural catastrophe risk. Option 3: Federal Lending to State Catastrophe Funds; This option would create a federal lending facility to provide temporary loans at market prices to state catastrophe funds. Advantages: * This option could help state catastrophe insurance funds with financing needs after a disaster; * The federal lending option would require states to demonstrate that they were doing all they could to attract private capital. Disadvantages: * The federal lending option imposes credit risk on taxpayer– the risk that the loans would not be repaid; * Political pressure could be exerted to keep the terms and conditions of federal loans more favorable than those in the private market. Option 4: Insurance Company Catastrophe Reserving; This option would permit private insurance companies to establish tax-deferred reserves for future natural catastrophes. Advantages: * With reserves, insurance companies could be more willing to underwrite policies; * Insurance regulators could be more willing to approve risk-based rates for consumers, because premium income could be set aside in a reserve fund. Disadvantages: * Allowing insurance companies to build reserves could involve tax benefits that favored one type of activity over another and could hamper economic efficiency; * Reserves could be costly for the federal government, because they would reduce federal tax revenue. Option 5: Homeowner Catastrophe Savings Accounts; This option would permit individuals to establish tax-deferred reserves to pay expenses related to disasters. Advantages: * Allowing homeowners to use tax-deferred dollars to pay for catastrophe insurance could induce more people to buy it; * This option might encourage more homeowner mitigation activities. Disadvantages: * Such accounts may not be enough to induce people to buy costly catastrophe insurance and, thus, may not broaden citizen participation in natural catastrophe insurance programs; * These accounts would reduce federal tax revenues but must be weighed against any reduction in post-disaster spending by the federal government. Option 6: Favorable Tax Treatment for Catastrophe Bonds; This option would facilitate the onshore creation of catastrophe bonds through tax exemptions for income from the underlying assets. Advantages: * Favorable tax treatment of catastrophe bonds would increase the ability of insurance markets to access capital markets; * Insurance companies could be more willing to underwrite catastrophe risk because the risk could be passed on to investors. Disadvantages: * This option creates a new class of reinsurer that would operate under regulatory and tax advantages not afforded U.S. reinsurance companies; * It is not clear how this option would encourage risk-based premiums or lead to more citizen participation in catastrophe insurance programs. Option 7: Property Tax Assessment for Private Insurance with Federal Deductible Payment; This option has property tax assessments paying a premium for an all-perils catastrophe insurance policy that would be provided by private insurance companies, with the federal government responsible for the deductible. Advantages: * This option would protect the tax base of a state’s economy; * The property tax assessment option would increase homeowner participation in catastrophe insurance programs. Disadvantages: * The property tax assessment option would reduce federal tax revenue; * This option could be expensive for taxpayers because the federal government would pay some portion of the homeowner deductible. Source: GAO. What GAO Recommends: This report does not contain recommendations. However, GAO evaluates seven public policy options that are discussed in the right-hand column and on the reverse side of this page. In written comments, the National Association of Insurance Commissioners (NAIC) generally agreed with GAO’s report findings. To view the full product, including the scope and methodology, click on [hyperlink, http://www/GAO-08-07]. For more information, contact Orice Williams at (202) 512-8678 or williamso@gao.gov. [End of section] Contents: Letter: Results in Brief: Background: Government Natural Catastrophe Insurance Aims to Provide Affordable Protection but Often Requires Postfunding after Large Natural Catastrophes: Because Many Americans Are Inadequately Insured for Natural Catastrophes, Federal Programs Play a Significant Role in Recovery: Options for Changing the Federal Role in Natural Catastrophe Insurance Attempt to Address Market Issues but May Not Limit Federal Exposure: Agency Comments and Our Evaluation: Appendixes: Appendix I: Objectives, Scope, and Methodology: Appendix II: Selected State Natural Catastrophe Insurance Programs: Appendix III: Comments from the National Association of Insurance Commissioners: Appendix IV: GAO Contact and Staff Acknowledgments: Related GAO Products: Figures: Figure 1: Catastrophic Risk in the United States--Earthquake, Hurricane, Tornado, and Hail: Figure 2: Insured Property Value, by County: Figure 3: Estimated Federal Disaster Assistance Obligated for Homeowners and Renters Lacking Adequate Insurance in Five Gulf Coast States after the 2005 Hurricane Season: Figure 4: Advantages and Disadvantages of Seven Public Policy Options for Changing the Federal Role in Natural Catastrophe Insurance: Figure 5: Comparison of Selected State Natural Catastrophe Insurance Programs in 2005: Abbreviations: CDBG: Community Development Block Grant: CEA: California Earthquake Authority: DLP: Disaster Loan Program: FAIR: Fair Access to Insurance Requirements: FCIC: Federal Crop Insurance Corporation: FEMA: Federal Emergency Management Agency: FHCF: Florida Hurricane Catastrophe Fund: FIRM: Flood Insurance Rate Map: FWUA: Florida Windstorm Underwriting Association: GUA: Georgia Underwriting Association: HA: Housing Assistance: HRA: High Risk Account: HUD: Department of Housing and Urban Development: IHP: Individuals and Households Program: JUA: Florida Residential Property and Casualty Joint Underwriting Association: MIUA: Mississippi Insurance Underwriting Association: NAIC: National Association of Insurance Commissioners: NFIP: National Flood Insurance Program: NOAA: National Oceanic and Atmospheric Administration: ONA: Other Needs Assistance: PLA: Personal Lines Account: SBA: Small Business Administration: SFHA: Special Flood Hazard Areas: [End of section] United States Government Accountability Office: Washington, D.C. 20548: November 26, 2007: The Honorable Spencer Bachus: Ranking Member: Committee on Financial Services: House of Representatives: Dear Mr. Bachus: Natural disasters not only devastate communities and individuals but also are costly to insurers, state governments, and the federal government. As the 2005 hurricane season showed, costs associated with a large natural disaster, such as a hurricane or major earthquake, can be enormous. Aside from the human toll, Hurricanes Katrina, Rita, and Wilma caused insured losses of an estimated $56.5 billion and total economic losses--including both insured and uninsured losses--of more than $100 billion. The federal government alone has appropriated more than $88 billion for disaster relief and recovery from the 2005 hurricanes. Future disasters could be even more expensive. One catastrophe modeling firm estimates that a large hurricane in southeast Florida could cause insured losses of more than $130 billion and a total economic loss of more than $260 billion.[Footnote 1] Similarly, if San Francisco's 1906 earthquake were to be repeated today, it is estimated that it would cause insured losses of more than $70 billion and a total economic loss of more than $280 billion. Also, a rupture of the Puente Hills fault in the Los Angeles basin could lead to estimated insured losses of more than $140 billion and a total economic loss of more than $500 billion. As we reported earlier in 2007, large losses associated with natural catastrophes are some of the biggest exposures that property and casualty insurers face.[Footnote 2] To remain financially solvent, these companies must estimate and prepare for the potential impact of such events. Because of the increased risk of natural catastrophes and the challenges of predicting losses from them, insurance companies have started limiting both renewals of existing contracts and the writing of new contracts in catastrophe-prone regions of the United States, such as coastal areas (from Massachusetts to Texas), earthquake zones in California, and the Mississippi River area of the central United States along the New Madrid earthquake fault. Such events place enormous stress on insurance markets and governments, carry huge costs, and have raised concerns about who ultimately bears the costs and receives the benefits of government disaster insurance programs. For these reasons, debate has arisen about the appropriate role for the federal government in insuring against and in recovering from natural catastrophes.[Footnote 3] While many public policy observers agree that the federal government does and should play an integral role in disaster relief and infrastructure recovery, some other public policy observers have asked whether the government's current role is the most appropriate and have suggested alternatives. Some have argued for more federal involvement, but others believe that the federal government may be doing too much, crowding out private insurance and reducing the private market's ability and willingness to provide insurance-based solutions.[Footnote 4] Public policy observers have raised moral hazard concerns, noting that generous federal disaster relief may discourage homeowners from purchasing natural catastrophe insurance.[Footnote 5] These observers have also pointed out that government catastrophe insurance programs are vulnerable to adverse selection, in that homeowners who are at the most risk are also the most likely to buy catastrophe insurance.[Footnote 6] In the context of your concerns about finding ways to ensure that the federal government and the private sector each play an appropriate role in the provision of natural catastrophe insurance, we examined (1) the rationale for and resources of the federal and state programs that have supplemented or substituted for private natural catastrophe insurance and their funding; (2) the extent to which Americans living in catastrophe-prone areas of the United States are uninsured and underinsured and the federal payments that have been made to such individuals since the 2005 hurricanes; and (3) public policy options for revising the federal role in natural catastrophe insurance markets. We reviewed or analyzed documents on federal and state catastrophe insurance programs, the numbers of uninsured and underinsured and federal payments that have been made, options to redefine the federal role in natural catastrophe insurance, and goals that could be used as a basis for designing and evaluating options for change. We interviewed officials from public interest groups, insurance companies, reinsurance companies, insurance agents, insurance brokers, insurance and reinsurance associations, insurance agent associations, state catastrophe insurance and reinsurance plans, state insurance departments, state economic development departments, the National Association of Insurance Commissioners (NAIC), the Federal Emergency Management Agency (FEMA), the National Flood Insurance Program (NFIP), the Department of Agriculture's Risk Management Agency, the Department of Housing and Urban Development (HUD), the Small Business Administration (SBA), Fannie Mae, Freddie Mac, rating agencies, risk modeling organizations, academia, law firms, a hedge fund, an insurance research organization, a private research organization, a consumer group, and others. To determine the mechanisms that governments use to supplement or substitute for private catastrophe insurance markets, we collected oral and documentary information from public and private officials in various states with high and low catastrophe risk and from Washington, D.C. To determine the number of uninsured and underinsured Americans and payments made to such individuals after the 2005 hurricanes, we collected information from states, examined federal agency data, interviewed federal officials who prepared these data, sought information from the private sector, and interviewed state officials responsible for disbursing federal disaster funds. Data on the numbers and amounts of money going to the uninsured and underinsured were incomplete and had a number of limitations, which are described in appendix I. We determined that these data were sufficient for the purposes of this engagement. We identified various options for altering the role of the federal government in catastrophe insurance by looking at bills before the current and previous Congresses as well as other options that were not included in current legislative proposals--for example, a proposal before a committee of NAIC. After fieldwork for this report concluded, we were informed that additional public policy options not considered in this report were being discussed before a committee of NAIC. We sought out both supporters and critics of each option, and our discussion of the third objective presents mainly advantages and disadvantages that they have identified. We developed a four-goal framework that was based on challenges faced by current government natural catastrophe insurance programs and used the framework to analyze current options for changing the federal role in natural catastrophe insurance. We developed these goals by drawing insights from the following: past GAO work, legislative histories of laws that changed the roles of state governments and the federal government after disasters, bills before the current and previous Congresses, interviews with public and private sector officials, and refereed articles written by academics in insurance economics. Although we identified numerous possible goals that could assist our analysis, we believe the four goals that we chose accurately capture the essential concerns of the federal government. The congressional policy choices ahead involve striking an appropriate balance among these goals. The scope of our work covered hurricane and earthquake perils. We did not investigate tornado, hail, or other perils, such as wildfires. We focused on the property and casualty insurance line, especially homeowners insurance. In reporting on the amount of federal disaster assistance to individuals who lacked adequate insurance in the Gulf Coast states following the 2005 hurricanes, we attempted to identify payments to homeowners only. However, because in some instances we could not separate out payments to renters, we included payments to both homeowners and renters in some of our calculations. Judgments about whether market failure did or did not exist in any particular state's property and casualty insurance market--and whether the cost of doing something about it was or was not more expensive than the federal government's not acting--were outside the scope of this report. Appendix I contains additional details of our objectives, scope, and methodology. We did fieldwork in Alabama; California; Connecticut; Florida; Illinois; Indiana; Louisiana; Massachusetts; Mississippi; Missouri; New York; Ohio; Texas; and Washington, D.C. Our work was conducted between March 2006 and October 2007 in accordance with generally accepted government auditing standards. Results in Brief: The federal government and some states have developed natural catastrophe insurance programs that supplement or substitute for private natural catastrophe insurance. For example, the Federal Crop Insurance Corporation (FCIC) currently insures crops for losses from multiple perils, and NFIP insures against flood losses. Although these programs were created to provide affordable insurance coverage, by design they are not adequately funded--that is, the premium rates do not cover the government's exposure--and rely on postfunding mechanisms to cover catastrophic loss years. Unlike private insurers that base premium rates on the risk of loss associated with properties, these programs offer legislatively mandated premium subsidies to encourage participation, and Congress appropriates funds for emergency disaster relief as needed. Similarly, some state governments have intervened when private sector insurance became prohibitively expensive or was not widely available, offering state-sponsored catastrophe insurance programs. For example, California created an earthquake fund in 1996 when private insurers significantly reduced the writing of homeowner earthquake coverage following the Northridge Earthquake. Likewise, Florida has created the Citizens Property Insurance Corporation (Florida Citizens)--the largest home insurer in Florida--to provide state-backed insurance coverage, including for wind damage, for homeowners who cannot get coverage in the private sector. The natural catastrophe insurance programs in Florida, Louisiana, Texas, and other states are funded through a combination of premium payments and postevent assessments and bonds. Like the federal programs, some state natural catastrophe insurance programs have been criticized for not charging premiums sufficient to cover risks. After the 2005 hurricanes, for example, some of these programs faced large accumulated deficits and required substantial public funding to continue operations. The 2005 hurricanes made clear that, even with the federal and state natural catastrophe insurance programs, significant numbers of Americans lacked adequate insurance against natural catastrophes for their homes. These property owners were either uninsured or underinsured, for a variety of reasons. Perhaps most significantly, buying natural catastrophe insurance is in many cases voluntary, and homeowners may choose not to buy it because they do not understand their risk exposure, do not understand the protection catastrophe insurance offers, or cannot afford it. In some cases, homeowners have insurance, but it covers less than the full replacement value of their property or has other policy limitations. Underinsurance can be exacerbated following a natural catastrophe, when rebuilding costs can increase substantially. Uninsured and underinsured homeowners may compound the challenge of providing affordable natural catastrophe insurance by relying on the federal government for postdisaster assistance to rebuild their homes. These homeowners may seek federal disaster relief from several federal agencies, including grants from FEMA and HUD, and real property loans from SBA. As we found, a significant portion of post-Katrina payments to Americans have gone to homeowners who were inadequately insured. We estimated that a quarter to a third of all federal emergency appropriations after the 2005 hurricanes, or around $26 billion in grants and loans, was obligated to homeowners and renters who lacked adequate natural catastrophe insurance. As Congress and the industry continue to reevaluate the role of the federal government in insuring for natural catastrophes, Congress is faced with balancing the often-competing goals of limiting taxpayer exposure and ensuring that citizens are protected. We identified seven public policy options for changing the role of the federal government in natural catastrophe insurance, including a mandatory all-perils homeowners insurance policy, federal reinsurance for state catastrophe funds, a federal lending facility for state catastrophe funds, and several tax-based incentives to encourage greater participation by insurers and homeowners in managing natural catastrophe risks. As shown in figure 4, each of these options has advantages and disadvantages. As part of our evaluation, we weighed each of the options against four public policy goals that we identified for federal involvement in natural catastrophe insurance programs: (1) to have premium rates fully reflect actual risks, (2) to encourage private markets to provide natural catastrophe insurance, (3) to encourage broad participation in natural catastrophe insurance programs, and (4) to limit costs to taxpayers before and after a disaster. We found that a mandatory all- perils policy, for example, could help create broad participation and provide a private sector solution. But this option could also require subsidies for low-income residents and thus potentially create substantial costs for the federal government. Similarly, while federal reinsurance for state catastrophe funds could lead to greater participation by private insurers, it could also displace the private reinsurance market. Also, a federal lending facility could also help state catastrophe insurance funds with financing needs after a catastrophe but could also expose the federal government--and taxpayers--to the risk that a loan might not be repaid. Given the often- competing purposes of many public policy options, some options may be more appealing than others, but all warrant discussion as part of the current debate. While some options would address the goals of charging rates that reflect the true risk of catastrophic loss, encourage broad participation, or promote greater private sector participation, these goals must be balanced with the desire to make rates affordable. We provided a draft of this report to NAIC and provided excerpts from the draft to Alabama Insurance Underwriting Association (Alabama Beach Pool), California Earthquake Authority (CEA), Department of Housing and Urban Development (HUD), Federal Crop Insurance Corporation (FCIC), Federal Emergency Management Agency (FEMA), Florida Hurricane Catastrophe Fund (FHCF), Florida Citizens Property Insurance Corporation (Florida Citizens), Georgia Underwriting Association (GUA), Louisiana Citizens Property Insurance Corporation (Louisiana Citizens), Mississippi Windstorm Underwriting Association (Mississippi Windpool), the North Carolina Insurance Underwriting Association (North Carolina Beach Plan), Small Business Administration (SBA), the South Carolina Wind and Hail Underwriting Association (South Carolina Windpool), and the Texas Windstorm Insurance Association (Texas Windpool). NAIC provided written comments that are reprinted in appendix III. In these comments, NAIC officials said that our approach was thorough and that we had done an admirable job of evaluating the public policy aspects of the seven proposals. The officials also mentioned two additional proposals that merit consideration, including a proposal that includes an allocation system for determining what portion of hurricane damages should be attributed to wind and what portion to flood, and another proposal for the creation of a federal entity to oversee property insurance rates in coastal zones. While these options were put forth too recently to be included in our review and analysis, we will collect additional information about them, to the extent possible, during the course of other ongoing work involving NFIP. NAIC also commented on allegations made by some critics of state regulation, who have suggested that regulators may be suppressing rates for some catastrophe insurers, and cautioned against assuming widespread rate suppression by state insurance regulators. As NAIC noted, our report does not allege any such activity on the part of state insurance regulators but notes that it is a concern raised by some critics. We recognize that determining the appropriate rates for natural catastrophe insurance is challenging and is often a negotiated process between the insurers and regulators. Alabama Beach Pool, the CEA, FCIC, Florida Citizens, FHCF, the GUA, Louisiana Citizens, MWUA, the North Carolina Beach Plan, SBA, the South Carolina Windpool, and the Texas Windpool provided technical comments that we incorporated in this report as appropriate. Background: The United States is exposed to several major hazards, in particular earthquakes and hurricanes, in coastal areas. As shown in figure 1, the Pacific, South Atlantic, and Gulf Coasts face the highest risk of earthquakes and hurricanes. According to the National Oceanic and Atmospheric Administration (NOAA), 53 percent of the nation's total population, or approximately 153 million people, lived in coastal counties in 2003.[Footnote 7] Moreover, the total coastal population increased by 33 million people, or 28 percent, between 1980 and 2003. California led in coastal population change, with the number of residents increasing by 9.9 million people. Florida showed the greatest percentage population change between 1980 and 2003, increasing nearly 75 percent. The nation's coastal population is expected to increase by more than 7 million people by 2008 (over current levels) and by 12 million people by 2015. Figure 1: Catastrophic Risk in the United States--Earthquake, Hurricane, Tornado, and Hail: [See PDF for image] This figure is a map of the United States with depictions of the paths and/or landfalls of earthquakes, hurricanes, tornadoes and hail. Shaded areas on the map indicate risk, from low risk to high risk. Twenty-one different shadings are indicated on a continuum from lowest risk (1) to highest risk (21), with the following cities depicted: Phoenix: 2; Las Vegas: 3; New York: 4; Philadelphia: 5; Portland: 6; Washington, DC, Boston, Chicago: 7; San Diego: 9; Salk Lake City: 10; Raleigh: 11; Atlanta, San Antonio: 12; St. Louis, Denver: 13; Seattle, Los Angeles: 14; Anchorage: 16; Tampa, Houston, Dallas: 17; New Orleans, Memphis, Oklahoma City: 18; Charleston, San Francisco: 19; Honolulu: 20; Miami: 21. Source: Risk Management Solutions. [End of figure] The housing supply in coastal areas also continues to grow, despite the high risk of earthquakes and hurricanes. NOAA reported that coastal counties contained 52 percent of the nation's total housing supply in 2000.[Footnote 8] The leading states in terms of total housing units in coastal counties were California, Florida, and New York, which together have 41 percent of the total housing supply in these counties. One study put the estimated insured value of coastal property in states bordering the Atlantic Ocean and Gulf of Mexico at $7.2 trillion as of December 2004.[Footnote 9] As shown in figure 2, properties along the Pacific and North-Atlantic Coasts and the Gulf of Mexico have some of the highest insured property values. The value of residential and commercial coastal property in Florida and New York was $1.94 trillion and $1.90 trillion, respectively, in 2004. Figure 2: Insured Property Value, by County: [See PDF for image] This figure is a map of the continental United States, depicting the insured property value, by county. Counties are shaded into the following five categories by insured property value, in billions: Less than 5; Five to ten; Eleven to fifty; Fifty-one to one hundred; More than 100. Source: Copywrite, 2007 AIR Worldwide Corporation. [End of figure] Private Natural Catastrophe Insurance: Insurance coverage against natural catastrophes for a home may or may not be included in homeowners insurance contracts. For example, coverage against wind loss from an event such as a hurricane is typically included. However, in some areas of certain states--mostly coastal regions--wind coverage may be excluded from homeowners insurance contracts and may be available only through the surplus lines insurance market or a state-managed entity.[Footnote 10] Similarly, earthquake coverage is commonly excluded from homeowners insurance contracts and instead is sold separately by insurance companies or, in the case of California, through a state-managed program. The price of property and casualty insurance is affected by both the annual expected loss and the cost of diversifying the risk of catastrophic losses.[Footnote 11] Insurers can diversify the risk of catastrophic losses by, among other things, purchasing reinsurance, which is insurance for insurance companies, or by selling financial instruments such as catastrophe bonds. Insurance companies do not know in advance what their actual costs are going to be, because they can determine these costs only after a policy has expired. The insurer's objectives are to calculate premiums that will make the business profitable, enable the company to compete effectively with other insurers, and allow the company to pay claims and expenses as they occur. When insurers, reinsurers, and investors in catastrophe financial instruments perceive that the expected frequency or severity of natural catastrophes has increased, they may increase the price of insurance. If a company believes that the risk of loss--for example, from flooding or earthquake--is unacceptably high given the rate that can be charged, it declines to offer coverage. State Regulation of Insurance Prices: While the federal government retains the authority to regulate insurance, it has given primary responsibility for insurance regulation to the states, in accordance with the McCarran-Ferguson Act of 1945.[Footnote 12] State insurance commissioners are responsible for regulating rates, monitoring the availability of insurance, and assessing insurance firms' solvency. The insurance regulators of the 50 states, the District of Columbia, and the U.S. territories have created NAIC to coordinate regulation of multistate insurers. NAIC serves as a forum for the development of uniform policy, and its committees develop model laws and regulations that, when adopted by state legislatures or promulgated by state regulators, govern the U.S. insurance industry. Critics of state insurance regulation argue that insurance prices and terms of coverage, particularly for homeowners insurance in areas prone to natural catastrophes, are highly regulated and that the insurance industry is generally not allowed to respond freely to changing risks or market conditions. In particular, these critics say that: * insurance regulators do not allow private insurers in catastrophe- prone areas to charge rates sufficient to build surpluses or transfer risks to reinsurers, * regulators may be subject to voter pressure and thus to legislative pressure to keep insurance premiums affordable and coverage readily available, and: * regulatory and political restrictions prevent markets from giving consumers accurate price signals regarding the risks of living in catastrophe-prone areas. NAIC officials told us that projected loss costs to cover the insurer's catastrophe exposure vary widely depending on which risk-modeling firm the insurer selects to produce its catastrophe loss costs. Only future results prove whether insurance company actuaries or insurance regulator actuaries are correct. The officials said that one should not assume that insurers and their actuaries have perfect information about what catastrophes will occur during the next year and about how the economy will behave. They added that one should also not assume that actuaries working for insurance companies are always correct in their projections of the needed price for the future experience period and that actuaries working for insurance regulators are always wrong. In the aftermath of natural catastrophes, some insurers responded by limiting their exposure in catastrophe-prone areas with restrictions on underwriting, higher deductibles, and lower coverage limits.[Footnote 13] In particular, there were property insurance affordability and availability crises in the Gulf Coast states or Florida after Hurricane Camille in 1969, Hurricane Celia in 1970, Hurricane Andrew in 1992, and the 2005 hurricanes; and in California following the Northridge Earthquake in 1994. Various proposals have been put forth over the past 15 years seeking to have the federal government take a larger role--for example, as a reinsurer or by allowing insurance companies to accumulate tax-deferred reserves--in addressing the affordability and availability of natural catastrophe insurance.[Footnote 14] Federal Natural Catastrophe Insurance and Disaster Aid: The federal government engages in a wide variety of insurance activities, among them providing multiperil crop insurance to farmers and flood insurance to homeowners and businesses.[Footnote 15] In addition, the federal government provides disaster assistance to individuals and households.[Footnote 16] FEMA, SBA, and HUD are the primary agencies administering federal disaster relief and recovery programs. FCIC provides insurance coverage for farmers who suffer financial losses when their crops are damaged by droughts, floods, or other natural disasters. By law, FCIC pays the premium for catastrophic coverage against losses of 50 percent of a farm's normal yield at 55 percent of the market price.[Footnote 17] In addition, FCIC offers premium subsidies for "buy-up" coverage against crop, revenue, and prevented planting losses, with coverage for losses ranging from 50 to 90 percent of a farm's normal yield.[Footnote 18] FCIC estimates that participation of eligible farmers is approximately 80 percent of acres planted. FEMA, through NFIP, offers insurance to homeowners and businesses for losses due to flooding and currently has 5.3 million policyholders. By law, NFIP must offer reduced premium rates for homes built in floodplains prior to the creation of flood insurance rate maps (pre- Flood Insurance Rate Map (FIRM) properties). About one quarter of NFIP policies are pre-FIRM and pay about 40 percent of the risk-based rate. According to NFIP, homes built in floodplains to an approved building code after the creation of flood maps pay actuarially sound premiums. Participation in the program is mandatory for homeowners with mortgages issued by federally regulated lenders on properties in special flood hazard areas (SFHA) where flood insurance is available.[Footnote 19] According to the RAND Corporation, about half of all homeowners who live in SFHAs purchase flood insurance. In addition to providing crop, flood, and other insurance, the federal government provides disaster assistance to individuals. FEMA provides disaster relief and recovery assistance to individual citizens through its Individuals and Households Program (IHP), which is intended to provide money and services to people in a disaster area when losses are not generally covered by insurance and property has been damaged or destroyed. IHP includes Housing Assistance (HA) and Other Needs Assistance (ONA). FEMA may provide five types of HA: financial assistance to rent temporary housing, "direct" temporary housing assistance, repair assistance, replacement assistance, and permanent housing construction in certain areas outside of the continental United States and other remote areas. FEMA may provide ONA grant funding for transportation expenses, medical and dental expenses, and funeral and burial expenses. ONA grant funding may also be available to replace personal property, repair and replace vehicles, and reimburse moving and storage expenses under certain circumstances. IHP is not intended to restore damaged property to its predisaster condition. SBA's Disaster Loan Program (DLP) is the primary federal program for funding long-range recovery for private sector, nonfarm disaster victims. Eligible losses include under or uninsured damages and can not duplicate benefits received from another source (i.e. insurance recovery, FEMA, etc.) The Small Business Act authorizes SBA to make available the following two types of disaster loans: (1) physical disaster home loans to homeowners, renters, and businesses of all sizes, and (2) economic injury disaster loans to small businesses.[Footnote 20] Homeowners and renters can borrow up to $40,000 for repair or replacement of household and personal effects. Homeowners can also borrow up to $200,000 to repair or replace a primary residence. Businesses of all sizes can borrow up to $1.5 million to repair or replace disaster damaged real estate, machinery and equipment, inventory, etc. Small businesses can borrow up to $1.5 million for disaster related economic injury resulting from the declared disaster. The combined loans to a business for physical loss and economic injury cannot exceed $1.5 million. Homeowners and businesses must provide reasonable assurance that they can repay the loan out of personal or business cash flow, and they must have satisfactory credit and character. HUD also provides disaster recovery assistance through several programs. After the 2005 hurricanes, Congress appropriated $16.7 billion to the Community Development Block Grant (CDBG) program for disaster recovery. The CDBG program generally provides funding to metropolitan cities and urban counties that have been designated as entitlement communities and to states for distribution to other communities.[Footnote 21] Grant recipients must give maximum feasible priority to activities, including emergency-related activities, that (1) benefit low-and moderate-income families or aid in the prevention or elimination of slums or blight, or (2) meet urgent community development needs. However, HUD can waive regulatory and statutory program requirements to increase the flexibility of the CDBG funds for disaster recovery. These grants afford states and local governments a great deal of discretion to help them recover from presidentially declared disasters.[Footnote 22] Government Natural Catastrophe Insurance Aims to Provide Affordable Protection but Often Requires Postfunding after Large Natural Catastrophes: Government natural catastrophe insurance programs were created because certain perils are difficult to insure privately and because, when private insurance is available, it may not be affordable. To keep natural catastrophe insurance available and affordable, government insurance programs operate differently than private insurance companies. Private insurance companies generally rely on premiums collected from those they insure to cover operating costs and losses and set premium rates at levels that are designed to reflect the risk that the company assumes in providing the insurance. These companies may also accumulate reserves to cover large losses. Federal and state government insurance programs also collect up-front premiums, but their rates do not always reflect the risks that the programs assume. Because premiums are inadequate to cover operating costs and losses, the government programs generally have limited resources and often face deficits after disasters. However, unlike private insurers, federal insurers may obtain funds after a catastrophic event through emergency appropriations. State programs may also access postevent funding through various means, including assessments on private insurers, bonds, and private reinsurance. State programs may also be postfunded through state general revenue funds and federal disaster relief payments. This structure has several implications. First, it may encourage homeowners in catastrophe-prone locations to seek coverage from government programs, crowding out the private market and increasing the government's financial exposure. Second, homeowners may not receive appropriate price signals about the risk of living in catastrophe-prone locations. Third, taxpayers who live in less risky locations may be subsidizing those living in catastrophe-prone locations. Finally, the added burden of private insurers' assessment obligations may provide another reason for them to leave already stressed markets. Federal natural catastrophe insurance programs fill gaps in private insurance markets and help limit disaster relief payments. For example, FCIC and NFIP were created because private insurers had determined that multiperil crop and flood losses were uninsurable and declined to provide coverage. A 1937 study by the Executive Committee on Crop Insurance, which noted that commercial attempts to insure against crop losses had been unsuccessful, provided the impetus for creating FCIC in 1938. Initially, the program was experimental and suffered heavy losses. The Federal Crop Insurance Act of 1980 expanded the program to replace free disaster coverage (in the form of compensation to farmers who were unable to plant crops and who suffer yield losses) with insurance. The flood insurance program was initiated because it had become clear by the 1950s that private insurance companies could not profitably provide affordable flood coverage because of the catastrophic nature of flooding and the impossibility of developing an actuarial rate structure that could adequately reflect the risk to flood-prone properties, among other reasons. One of the primary purposes of the National Flood Insurance Act of 1968, which created NFIP, was to reduce federal expenditures for disaster assistance and flood control. State natural catastrophe insurance programs were created to avoid homeowners insurance crises that threatened the states' housing markets. For example, the California Earthquake Authority was formed in 1996 in response to a crisis in the residential property insurance market following the Northridge earthquake in 1994. According to the Insurance Information Institute, California insurers had collected only $3.4 billion in earthquake premiums in the 25-year period prior to the Northridge earthquake but had paid out more than $15 billion on Northridge claims alone. Moreover, insurers representing about 95 percent of the homeowners insurance market in California began to limit their exposure to earthquakes by writing fewer or no new homeowners insurance policies, triggering a crisis that threatened California's housing market and stalled the state's recovery from recession. See appendix II for a more detailed description of state natural catastrophe insurance programs. Florida Citizens is a nonprofit tax-exempt entity that provides residential and commercial property insurance coverage when private insurance is not available. Florida Citizens was established in 2002 after two separate insurance pools--the Florida Windstorm Underwriting Association (FWUA) and the Florida Residential Property and Casualty Joint Underwriting Association (JUA)--were combined.[Footnote 23] In addition, the Florida Hurricane Catastrophe Fund (FHCF) provides an alternative to traditional hurricane reinsurance, reducing the cost of coverage significantly below that of private reinsurance and lowering the cost of insurance to homeowners. The FHCF was established in 1993 in response to Hurricane Andrew, which resulted in a severe shortage of catastrophe property reinsurance capacity, stricter policy terms and conditions, and sharp increases in property catastrophe reinsurance rates in the year following the storm.[Footnote 24] The post-Andrew reaction of a number of insurance companies was to attempt to reduce their underwriting exposure, and 39 insurers stated in early 1993 that they intended to either cancel or not renew approximately 844,000 policies in Florida.[Footnote 25] Other states--including Alabama, Louisiana, Mississippi, and Texas--have created state funds to make natural catastrophe insurance available and affordable. Premium Rates for Government Natural Catastrophe Insurance Programs Are Often Determined by Factors Other Than Risk: Because government natural catastrophe insurance programs are often created to ensure the availability and affordability of natural catastrophe insurance, homeowner premiums for these programs--although risk-related--are generally not based entirely on the homeowners level of risk. Federal natural catastrophe insurance program premium rates are often set by statute and involve government subsidies. For example, to encourage broad participation in the crop insurance program, federal law seeks to ensure that the premiums are affordable to all farmers by requiring FCIC to pay a portion of the premium cost.[Footnote 26] Specifically, FCIC offers farmers varying subsidy rates for crop insurance, depending on the level of protection they seek.[Footnote 27] Crop insurance subsidies totaled about $2.3 billion in crop years 2005 and 2006. In addition, federal crop insurance legislation directs FCIC to operate at a loss ratio of no more than 1.075.[Footnote 28] A loss ratio greater than 1.00 indicates that the program paid more in claims than it collected in premiums.[Footnote 29] Furthermore, we have previously reported that NFIP is not designed to be actuarially sound.[Footnote 30] Annually, flood insurance subsidies total about $1.3 billion.[Footnote 31] State natural catastrophe insurance program premium rates may also be set by statute. Florida Citizens historically has been required to maintain premium rates that were not competitive with the private insurance market. However, in January 2007, the Florida Legislature allowed Florida Citizens to charge competitive rates.[Footnote 32] Even by 2006, Florida Citizens was the largest property insurer in Florida. It receives much of its reinsurance coverage from the FHCF, which charges premium rates that are estimated to be about a quarter to a third the cost of private market reinsurance. The program can charge these rates because of its tax-exempt status and ability to postfund claims losses through bonds, among other advantages. These two state programs are able to charge lower premiums than private insurance companies, encouraging more people to seek coverage in the state programs and leaving the state more financially vulnerable in the event of a large hurricane. State natural catastrophe insurance program premium rates are also subject to approval by state insurance regulators that have generally resisted rate increases. The Mississippi Windstorm Underwriting Association (Mississippi Windpool) provides coverage against windstorms and hail for people in the six coastal counties of Mississippi who might not be able to get wind coverage in the private insurance market. After Hurricane Katrina, the Mississippi Windpool sought a rate increase of almost 400 percent, primarily to cover the increased cost of reinsurance. The state insurance regulator granted a 90 percent increase. Furthermore, the state government will use $50 million in federal disaster recovery funds provided by HUD to offset the increased cost of reinsurance in 2007 and 2008. In addition, the state government created a reinsurance fund that uses state general revenue funds to offset the increased cost of reinsurance. Similarly, the Texas Windstorm Insurance Association (Texas Windpool) offers wind and hail coverage in 14 coastal counties and other specified areas. By law, Texas Windpool residential and commercial premium rates may not increase more than 10 percent above the rates for noncommercial windstorm or hail insurance that are in effect at the time the request for an increase is filed. However, the insurance commissioner may suspend this rule to ensure rate adequacy in the catastrophe area.[Footnote 33] In May 2006, the Texas Windpool sought a 19 percent residential and 24 percent commercial rate increase, but the insurance commissioner approved a 3.1 percent residential and 8 percent commercial rate increase. When the Texas Windpool sought a 20 percent residential and 22 percent commercial rate increase in November 2006, the insurance commissioner approved a 4.2 percent residential and 3.7 percent commercial rate increase. In both instances, the insurance commissioner stated that he favored an incremental approach to strengthening the Texas Windpool that did not put an undue economic burden on coastal homeowners. Federal and State Natural Catastrophe Insurance Programs Have Incurred Large Postdisaster Deficits Because of Inadequate Resources and Reliance on Postfunding: Unlike private insurance companies, government natural catastrophe insurance programs often do not employ accrual accounting and are not always required to accumulate adequate resources to meet their obligations. Generally, insurance premiums are paid in advance, but the period of protection extends into the future. Private insurers are required by statutory accounting rules to establish reserves for incurred or known claims and for the cost of "incurred but not reported" claims to ensure that the premiums collected in advance will be available to pay future losses.[Footnote 34] Incurred but not reported claims are insured losses that have already happened but that for any of a variety of reasons have not yet been reported to the insurer. Most government natural catastrophe insurance programs are not required to have these resources, because they are structured to postfund losses. As we have previously mentioned, NFIP and the federal crop insurance program are postfunded by emergency appropriations from federal taxpayers. State programs are generally postfunded by several mechanisms, including assessments on private insurers, bonds, and proceeds from general revenues. In most property and casualty insurance lines, state assessments are often passed through to policyholders. As a result, homeowners living in less risky locations also contribute to cover the shortfall--a scenario known as cross-subsidization. In those states where assessments cannot be passed through in some manner, private insurers must pay the assessments, while at the same time paying large claims from their own policyholders. In such instances, some companies may be reluctant to continue offering coverage in the state or may become insolvent. In the wake of recent natural catastrophes, some government natural catastrophe insurance programs suffered losses that eliminated their accumulated resources. For example, NFIP reported unexpended cash of approximately $1 billion following fiscal year 2004, but the program had suffered almost $16 billion in losses from Hurricane Katrina alone as of May 31, 2007.[Footnote 35] Similarly, Florida Citizens' high-risk account had a surplus of approximately $1.1 billion prior to the 2004 hurricane season, but the program incurred over $2 billion in losses from the 2004 hurricanes and almost $2 billion in losses from the 2005 hurricanes.[Footnote 36] The FHCF had accumulated net assets of $5.5 billion at the end of the 2004 fiscal year but had an estimated shortfall of approximately $1.4 billion following reimbursements to participating insurers after the 2004 and 2005 hurricane seasons.[Footnote 37] Prior to 2007, the Mississippi Windpool did not have resources beyond premiums and reinsurance because year-end profits and losses were shared by member companies. By the end of 2005, following Hurricane Katrina, the Mississippi Windpool had incurred a net loss of $473 million.[Footnote 38] In Louisiana, Citizens Property Insurance Corporation (Louisiana Citizens), which has a structure similar to that of Florida Citizens, had $80 million in cash reserves prior to the 2005 hurricane season but suffered more than $1 billion in losses after Hurricanes Katrina and Rita.[Footnote 39] Emergency appropriations authorizing funding for federal natural catastrophe insurance programs after disasters have often been significant. In the case of FCIC, not only are premium rates subsidized by almost 59 percent for the most popular coverage, but farmers may receive additional emergency disaster relief--for example, farmers received $1.6 billion following Hurricane Katrina.[Footnote 40] In the case of NFIP, not only are premium rates for pre-FIRM homes subsidized up to 60 percent on average, but after Hurricane Katrina NFIP was authorized to borrow over $20 billion to pay claims. State natural catastrophe insurance programs have also often required postfunding to satisfy their obligations in the wake of large natural catastrophes. For example, to fund its 2004 and 2005 deficits, Florida Citizens assessed insurance companies in most property and casualty lines $516 million and $205 million, respectively, and these amounts will be passed through to policyholders.[Footnote 41] In addition, the Florida Legislature appropriated $715 million from the general revenue fund to reduce the size of the 2005 deficit. Furthermore, to fund a bond issuance to cover the FHCF's shortfall, eligible Florida insurance policyholders incurred a 1 percent assessment that will be levied over at least 6 years beginning in January 2007. In June 2006, the FHCF issued a $1.35 billion postevent revenue bond to cover 2005 losses, and in July 2006 it issued a $2.8 billion preevent financing bond to provide liquidity for 2006 and future years. Similarly, Louisiana Citizens assessed all property insurance companies in the state $193 million after the 2005 hurricanes. It has also issued a postevent bond for $978 million to cover 2005 losses that will be financed by emergency assessments on insurers in certain lines of property and casualty insurance. These assessments are levied directly on policyholders, who may claim a tax credit against state income tax. The assessments will continue for as many years as needed to cover the plan's deficit. Both Florida Citizens and Louisiana Citizens have been declared to be municipalities rather than insurance companies by their respective state legislatures, and as a result cannot declare bankruptcy until the bond obligations are satisfied. In addition, the Mississippi Windpool funded its deficit through $525 million in assessments on member companies in proportion to their share of business in the state.[Footnote 42] At the time, these assessments could not be directly passed through to policyholders. At least one private insurance company found that its assessment liability was more than the entire amount of premiums it collected in the state and was forced to liquidate. Finally, the Texas Windpool assessed private insurance companies in Texas for the first $100 million in program losses and expenses from Hurricane Rita beyond its ability to pay from premiums and other income. Because Many Americans Are Inadequately Insured for Natural Catastrophes, Federal Programs Play a Significant Role in Recovery: The 2005 hurricanes illustrated how many Americans are uninsured and underinsured for natural catastrophes and the federal government's role in recovery from natural catastrophes. An analysis by HUD found that of the 192,820 owner-occupied homes with major or severe damage from Hurricanes Katrina, Rita, and Wilma, approximately 78,000, or about 41 percent, did not have any insurance or did not have enough insurance to cover the damage incurred.[Footnote 43] Homeowners do not purchase natural catastrophe insurance for a variety of reasons, including financial reasons. Moreover, buying a natural catastrophe insurance policy does not guarantee complete coverage for a dwelling. For example, if the home's replacement value is calculated inaccurately, the homeowner will buy too little insurance to cover all of the damage. More and more frequently, responsibility for supporting the needs of individuals who lack adequate insurance against natural catastrophe risk is falling to the federal government. We estimate that the federal government made approximately $26 billion available for homeowners and renters who lacked adequate insurance in response to the 2005 hurricanes. Homeowners May Not Be Insured against Natural Catastrophes for Several Reasons: Homeowners may not purchase natural catastrophe insurance because they face budget constraints, underestimate the risk they face, or fail to understand the protection such insurance affords.[Footnote 44] Information on the number of individuals who are uninsured against natural catastrophe risks is somewhat limited but helps demonstrate the extent to which homeowners do not purchase natural catastrophe insurance. About 41 percent of homes that sustained severe damage from any peril during the 2005 hurricanes were uninsured or underinsured. HUD reported that of the 60,196 owner-occupied homes with major or severe wind damage, almost 23,000, or 38 percent, lacked insurance against wind loss. Also, the Insurance Information Institute reported that about 86 percent of Californians did not have earthquake insurance on their homes in 2004. Furthermore, only about one half of eligible single-family homes in Special Flood Hazard Areas (SFHA) nationwide have purchased flood insurance.[Footnote 45] In areas outside of SFHAs, where flood insurance is voluntary, only about 1 percent of owners of single-family homes have purchased flood insurance, even though 20 to 25 percent of NFIP's claims come from outside of SFHAs. Purchasing insurance to protect homes against natural catastrophes is mandatory for some homeowners, but often it is voluntary. For example, homeowners who do not have mortgages are generally not required to have property and casualty coverage, and in some areas certain types of hazards are routinely excluded from homeowners policies. As we have seen, wind coverage is often excluded in some coastal areas, and the surplus lines market or a state-managed entity may offer coverage separately. Although lenders may require homeowners to purchase this supplemental insurance, those who own their homes outright may choose not to buy it. A similar situation exists with earthquake coverage in certain areas of the country. In earthquake-prone areas, earthquake coverage is commonly excluded from the homeowners insurance contract and is sold separately by insurance companies or, as in the case of California, by a state-managed program. In general, lenders do not require earthquake insurance as a condition of extending a mortgage. Consumers will purchase natural catastrophe insurance on the basis of their perception of risk. Studies have shown that consumers often consider the likelihood of a future catastrophe to be much lower than insurance companies' estimates. According to academic research, some homeowners may underestimate the risk of loss, have an overly optimistic view of expected losses, or be unaware that insurance is available. One insurance expert has concluded that if people believe that the chance of a serious event occurring is low, they often consider insurance unnecessary and will not seek out information on its benefits and costs. Reluctance to purchase insurance protection can be compounded by budget constraints. For some homeowners with relatively low incomes, disaster insurance is considered an expense that can be made only after taking care of necessities. An insurance expert has noted that insurance trade associations, consumer advocacy groups, and governments can provide better information to consumers about risk probabilities, insurer profitability, and prices to motivate better insurance purchasing behavior.[Footnote 46] One study of those living in earthquake zones has identified a variety of reasons for declining to purchase earthquake insurance.[Footnote 47] Some consumers are unwilling or reluctant to pay high premiums to insure against potentially large but rare disaster losses. Some consumers believe that the deductible for earthquake insurance--the standard deductible is 15 percent of the value of the home--is too high, given the premium rates and amount of coverage provided.[Footnote 48] A study of flood insurance market penetration rates cites several reasons why people do not purchase flood insurance.[Footnote 49] For property owners in SFHAs, the decision to purchase insurance is affected primarily by its price. Outside of SFHAs, property owners are not purchasing flood insurance because they may not be aware of flood risk, and because flood insurance agents have less interest in promoting flood insurance and in learning how to write flood policies. Also, certain limitations of the coverage, such as limits on basement flooding, make the policies less attractive in inland areas. Inaccurate Home Valuations Can Result in Underinsurance: Text box: Coinsurance Clause Formula: Homeowners who carry less than 80 percent of the full replacement cost of their homes receive an amount calculated by formulas. Generally, the amount will be the larger of the following two amounts: (1) Actual cash value of that part of the building damaged (minus depreciation cost) or; (2) Amount of insurance carried divided by 80 percent multiplied by replacement cost, then multiply that amount by Loss. To illustrate, assume that a dwelling has a replacement cost of $400,000, but it is insured for only $240,000. The roof of the house is 10 years old and has a useful life of 20 years, so it is 50 percent depreciated. Assume the roof is severely damaged by a hurricane, and the replacement cost of a new roof is $40,000. Ignoring the deductible, the insured receives the larger of the following two amounts: (1) Actual cash value equals $40,000 minus $20,000, which equals $20,000; or: (2) $240,000 divided by 80 percent times $400,000; multiplied by $40,000, equals $30,000. The insured receives $30,000 for the loss. The entire loss of $40,000 would have been paid if the insured had carried at least $320,000 of insurance. [End of text box] Homes may be underinsured because replacement costs are not calculated accurately. Replacement cost has been defined as the amount necessary to repair or replace the dwelling with material of like kind and quality at current prices. Replacement cost may not be calculated accurately for several reasons, including the effects of inflation, custom home building, remodeling, high demand for contractors, and changes in building codes following a natural catastrophe. Generally, property insurance losses are partial losses rather than total losses. However, in catastrophe-prone areas, the prospect of a total loss of property is real. If a homeowner suffers a total loss of property as a result of a natural catastrophe and the replacement cost has not been properly calculated, the property will not be fully insured. An insurance industry consultant estimates that in 2006 approximately 58 percent of the residential housing stock in the United States was undervalued for insurance purposes by an estimated 21 percent. Homeowners insurance coverage can vary by type of policy and from insurer to insurer, but there are fundamental similarities. The broadest coverage generally provides that a policyholder will receive full replacement cost with no deduction for depreciation (up to the policy limit) if a policyholder maintains coverage limits of 80 percent or more of the dwelling's full replacement cost.[Footnote 50] Otherwise, the homeowner receives a lesser amount according to the formula in the policy (see sidebar). The reasons that replacement costs may not be calculated accurately, leaving homeowners underinsured, are complex. First, replacement costs must be periodically updated to account for inflation. Second, beginning in the early 1980s developers began building more custom homes, and a significant percentage of homes were remodeled, sometimes extensively. Historically, the methodologies that the insurance industry used to calculate replacement costs did not always capture custom features. The industry has improved its calculation methodologies, but an insurance industry consultant told us that a large number of policies had not been properly updated. Furthermore, homeowners whose properties were remodeled may not have understood the need to tell their insurers about the remodeling, possibly to avoid rate increases. The problem of underinsurance can be exacerbated in the wake of a natural catastrophe when demand for contractors and materials to repair homes is high and the supply is tight. This phenomenon is known as "demand surge." In these circumstances, the short-term costs of repairing and rebuilding homes can escalate substantially, and replacement costs become significantly higher. In addition, over time a community may implement improved building codes, so that rebuilding may have to conform to stricter standards than those that were in place when a dwelling was first built.[Footnote 51] This situation can also make replacement costs much higher, as it did in Florida in the aftermath of Hurricane Andrew in 1992. Large Amounts of Federal Postdisaster Aid Have Been Distributed to Uninsured and Underinsured Homeowners: As of May 2007, Congress approved approximately $88 billion in emergency appropriations to assist in relief and recovery efforts in the Gulf Coast states following the 2005 hurricanes.[Footnote 52] Three federal agencies--FEMA, SBA, and HUD--received over $60 billion, or about two-thirds, of this amount. As we have previously noted, these agencies play a significant role in distributing federal disaster relief funds to individual victims. We estimate that, as of June 2007, the agencies had obligated approximately $26 billion, or between a quarter and a third, of the emergency appropriations to homeowners and renters in Alabama, Florida, Louisiana, Mississippi, and Texas who lacked adequate insurance (see fig. 3). Figure 3: Estimated Federal Disaster Assistance Obligated for Homeowners and Renters Lacking Adequate Insurance in Five Gulf Coast States after the 2005 Hurricane Season: [See PDF for image] The following data is depicted in table format: Hurricane-related emergency supplemental funding, in billions: Total: 87.755; Total congressional funds to assist Gulf Coast States: 100%. FEMA IHP: 42.576; SBA DLP: 0.988; HUD CDBG: 17.099; Portion of congressional funds to FEMA, SBA, and HUD: 69%; Estimated amount obligated by FEMA, SBA, and HUD to homeowners and renters lacking adequate insurance: FEMA IHP: 15.115; SBA DLP: 0.794; HUD CDBG: 9.904; Portion of congressional funds used for homeowners and renters who did not have adequate insurance: 29%. Source: GAO analysis of FEMA, HUD, and SBA documents and interviews. Note: These five Gulf Coast states are Alabama, Florida, Louisiana, Mississippi, and Texas. The 2005 hurricane season included Hurricanes Katrina, Rita, and Wilma. The SBA-obligated amount represents the subsidy cost of disaster loans at 14.64 percent, as of January 31, 2007. The FEMA-obligated amount includes HA, ONA, and manufactured housing funds as of June 18, 2007. The HUD-obligated amounts are as of May 16, 2007. The totals for each agency were calculated using different methodologies and data sources. For details, see appendix I of this report. [End of figure] Federal disaster assistance for homeowners and renters comes from FEMA, SBA, and HUD. For example: * For disasters declared between October 1, 2004, and October 1, 2005, FEMA could provide a maximum of $26,200 for housing and other needs assistance to an individual or household in a disaster area if property was damaged or destroyed and the losses were not covered by insurance.[Footnote 53] In total, FEMA obligated over $15 billion to homeowners and renters through IHP grants and manufactured housing. We have reported extensively on the difficulties that FEMA experienced in distributing disaster assistance through IHP.[Footnote 54] * Homeowners and renters can borrow up to $40,000 in personal property loans from SBA to repair or replace clothing, furniture, cars, and appliances damaged or destroyed in a disaster. SBA can also make real property loans up to a maximum of $200,000 to repair or restore a main residence to its predisaster condition. Any proceeds from insurance coverage on the personal property or home are deducted from the total loan amount. The interest rates on SBA disaster loans do not exceed 4 percent for those who are unable to obtain credit elsewhere or 8 percent for those who can get other credit. As of January 31, 2007, SBA approved over $5 billion in disaster loans for homeowners and renters after the 2005 hurricanes, at an interest subsidy cost of almost $800 million to the federal government. We have reported on the difficulties that SBA experienced in distributing disaster loans.[Footnote 55] * The largest recovery program for homeowners and renters after the 2005 hurricanes was HUD's CDBG program, which received $16.7 billion in supplemental appropriations to help homeowners with long-term recovery (including providing funds for uninsured damages), restore infrastructure, and fund mitigation activities in the declared disaster areas of Alabama, Florida, Louisiana, Mississippi, and Texas.[Footnote 56] To receive CDBG funds, HUD required that each state submit an action plan describing how the funds would be used, but the agency waived some program requirements for disaster recovery purposes. For example, HUD granted a waiver to Mississippi so that a portion of the CDBG funds could be used to pay reinsurance costs for 2 years for wind pool insurance maintained by the Mississippi Windpool. Two of the states receiving the largest allocation from the emergency CDBG appropriations were Louisiana and Mississippi, both of which opted to direct the vast majority of their housing allocations to homeowners. Both states based the amount of compensation that homeowners received on the value of their homes before the storms and the amount of damage that was not covered by insurance or other forms of assistance.[Footnote 57] The grants provided up to $150,000 for eligible homeowners.[Footnote 58] Both programs also attached various conditions to the acceptance of grants, such as requiring homeowners to rebuild their homes above the latest available FEMA advisory base flood elevation levels and establishing covenants to the land requiring that homeowners maintain hazard and flood insurance. It will be a challenge for federal, state, and local governments to sustain their current role in natural catastrophe insurance going forward. The Comptroller General of the Unites States has repeatedly warned that the current fiscal path of the federal government is "imprudent and unsustainable."[Footnote 59] In addition, we reported that, for state and local government sectors, large and growing fiscal challenges will begin to emerge within the next few years in the absence of policy changes.[Footnote 60] The fiscal challenges facing all levels of government are linked and should be considered in a strategic and integrated manner. Options for Changing the Federal Role in Natural Catastrophe Insurance Attempt to Address Market Issues but May Not Limit Federal Exposure: We identified seven public policy options for changing the role of the federal government in natural catastrophe insurance (see fig. 4). These policy options have many variants and are often contained in other proposals, including some bills that are before Congress.[Footnote 61] Some of these proposals are also being debated in venues such as the NAIC committees. We examined the advantages and disadvantages of these policy options and evaluated them against four broad public policy goals. These goals are: * charging premium rates that fully reflect actual risks, * encouraging private markets to provide natural catastrophe insurance, * encouraging broad participation in natural catastrophe insurance programs, and: * limiting costs to taxpayers before and after a disaster. Our analysis showed that each of the seven options met at least one of the policy goals but failed to meet others. The first option--a mandatory all-perils homeowners insurance policy--would help create broad participation and could provide a private sector solution. But this option could also require subsidies for low-income residents and thus potentially create substantial costs for the federal government that would have to be balanced against money saved from reduced disaster relief. A second option would involve providing federal reinsurance for state catastrophe funds--a change that could lead to greater private insurance market participation but that could also displace the private reinsurance market. A third option, establishing a federal lending facility for state catastrophe funds, could help such funds with financing needs after a catastrophe. But this option exposes the federal government to the risk that a state fund might not repay a loan and thus might not limit taxpayer exposure. The remaining four options include tax-based incentives to encourage greater participation by insurers and homeowners in managing natural catastrophe risks. These incentives offer some advantages, but could also represent ongoing costs to the federal government and taxpayers. Figure 4: Advantages and Disadvantages of Seven Public Policy Options for Changing the Federal Role in Natural Catastrophe Insurance: [See PDF for image] Option 1: All-Perils Homeowners Insurance Policy: This option would create a homeowner insurance policy that would provide coverage against all types of natural catastrophes. Advantages: * A mandatory all-perils policy would encourage broad participation in natural catastrophe insurance programs. * A mandatory all-perils policy could reduce the number of Americans needing postdisaster payments and possibly limit the federal government’s exposure. * A mandatory all-perils policy could eliminate the problems of uninsured property owners and adverse selection. * A mandatory all-perils policy would end homeowners’ uncertainty about coverage for some perils. Disadvantages: * It is not clear how the private market would be encouraged to underwrite all risks. * The all-perils option could require government subsidies for low- income property owners. * Premiums for an all-perils policy could be more expensive than current homeowner policy premiums, and these premium increases could be seen as unfair. * Enforcement of an all-perils policy could be extremely challenging. * Insurers traditionally oppose the all-perils option because of concern about large loss liabilities. Option 2: Federal Reinsurance for State Catastrophe Funds: This option would create federally backed reinsurance policies for state catastrophe funds. In one version of this option, states would create catastrophe funds that would be reinsured by the federal government. In another version, the Secretary of the Treasury would create an auction process for the sale of reinsurance contracts to private and state insurers and reinsurers. Advantages: * The federal reinsurance option could lead to greater participation from private insurers. * The federal reinsurance option would eliminate timing risk for insurance companies. * Insurance companies may be less interested in canceling policies in coastal regions if they have a stable source of reinsurance, their costs are reduced, and their liability is limited. * This option would not use tax dollars if the premium charged was risk- based. * The federal reinsurance option is preferable to federal disaster assistance. * This option would add stability to reinsurance rates. Disadvantages: * Federal reinsurance could compete with the private reinsurance sector. * Rates for federal reinsurance could be subject to consumer and hence political pressure to keep them below the private sector rates. * Federal taxpayer subsidies could favor those living in catastrophe- prone states. * Federal reinsurance could create inequities among states because of geographical differences in natural catastrophe risk. * Federal reinsurance could encourage further development and population growth in high-risk areas. * Government reinsurance that does not mimic what the private sector does could lead to government losses. Option 3: Federal Lending to State Catastrophe Funds: This option would create a federal lending facility to provide temporary loans at market prices to state catastrophe funds. Advantages: * This option could help state catastrophe insurance funds with financing needs after a disaster. * The federal lending option would eliminate timing risk for insurance companies. * Supporters of this option maintain that taxpayers would bear little or no insurance risk. * The federal lending option would require states to demonstrate that they were doing all they could to attract private capital, and this could lead to regulatory reform. Disadvantages: * It is not clear how this option would encourage risk-based premiums, would broaden citizen participation, or would be a cost-effective solution from the perspective of the federal government. * The federal lending option imposes credit risk on taxpayers—the risk that the loan would not be repaid. * The federal lending option could require the creation of a new federal agency to manage the program. * Political pressure could be exerted to keep the terms and conditions of federal loans more favorable than those in the private market. * This option would decrease incentives for insurers and reinsurers to accurately assess, underwrite, and price risk. Option 4: Insurance Company Catastrophe Reserving: This option would permit private insurance companies to establish tax- deferred reserves for future natural catastrophes. Advantages: * Tax-deferred reserving could mean that state regulators would be more willing to approve risk-based rates. * With reserves, insurance companies could be more willing to underwrite policies, thus encouraging a private sector solution. * This option would encourage broader-based citizen participation in catastrophe insurance programs. * Insurance regulators could be more willing to approve risk-based rates for consumers, because premium income could be set aside in a reserve fund. * Allowing insurance companies to build reserves could reduce pressure to create state catastrophe insurance programs. Disadvantages: * Allowing insurance companies to build reserves could involve tax benefits that favored one type of activity over another and could hamper economic efficiency. * Reserves could be costly for the federal government, because they would reduce federal tax revenue. * Tax-deferred reserves could be subject to manipulation, if they were used to smooth income flows across years and obscure current income. Option 5: Homeowner Catastrophe Savings Accounts: This option would permit individuals to establish tax-deferred reserves to pay expenses related to disasters. Advantages: * Allowing homeowners to use tax-deferred dollars to pay for catastrophe insurance could induce more people to buy it. * This option might encourage more homeowner mitigation activities. Disadvantages: * Such accounts may not be enough to induce people to buy costly catastrophe insurance and, thus, may not broaden citizen participation in natural catastrophe insurance programs. * These accounts would reduce federal tax revenues but must be weighed against any reduction of postdisaster spending by the federal government. Option 6: Favorable Tax Treatment for Catastrophe Bonds: This option would facilitate the onshore creation of catastrophe bonds through tax exemptions for income from the underlying assets. Advantages: * Favorable tax treatment of catastrophe bonds would increase the ability of insurance markets to access capital markets. * Insurance companies could be more willing to underwrite catastrophe risk because the risk could be passed on to investors. Disadvantages: * This option creates a new class of reinsurer that would operate under regulatory and tax advantages not afforded U.S. reinsurance companies. * It is not clear how this option is the most cost-effective for the federal government. * It is not clear how this option would encourage risk-based premiums or lead to more citizen participation in catastrophe insurance programs. * This option would only benefit larger insurers. * Given that catastrophe bonds were just issued by the two largest U.S. insurance companies, it is not clear why this tax treatment is needed. Option 7: Property Tax Assessment for Private Insurance with Federal Deductible Payment; This option has property tax assessments paying a premium for an all- perils catastrophe insurance policy that would be provided by private insurance companies, with the federal government responsible for the deductible. Advantages: * This option is market-based and designed to involve the private sector. • If risk-based premiums are required, the option is not a “government relief program.” * This option would protect the tax base of a state’s economy. * The property tax assessment option would increase homeowner participation in catastrophe insurance programs. * With this option, the high deductible could result in a lower insurance premium. Disadvantages: * The property tax assessment option would reduce federal tax revenue. * This option could be expensive for taxpayers because the federal government would pay some portion of the homeowner deductible. * It is not clear whether or not this option is a cost-effective solution for the federal government. * Premiums paid by homeowners might not be an effective signal of the risk of living in a particular location. * Homeowners could resist high property taxes implied by this option. Source: GAO. [End of figure] An All-Perils Policy Would Broaden Participation but Could Require Government Subsidies: A mandatory all-perils policy would require private insurers to provide coverage against all perils in a single standard homeowners policy that would be priced according to the risk of natural hazards each homeowner faced.[Footnote 62] For example, the policy would cover not only theft and fire but also wind, floods, and earthquakes. It would also be mandatory for all homeowners.[Footnote 63] This type of option offers several potential advantages. First, a mandatory all-perils policy, by definition, would encourage broad participation in natural catastrophe insurance programs. Moreover, including all American homeowners in natural catastrophe coverage could help reduce the number of Americans needing postdisaster payments and possibly limit the federal government's exposure. An all-perils policy would also eliminate existing gaps in coverage and remove the uncertainty many homeowners face in determining whether certain perils are covered and by whom--an issue that was spotlighted after Hurricane Katrina, when disputes emerged between private insurers and homeowners over the extent of the insurers' obligations to cover certain damages. Finally, because it would be mandatory and broad-based, an all-perils policy could lessen the problem of adverse selection that is often identified as the reason that some types of catastrophes, such as flooding, are considered to be uninsurable. This type of policy would spread risks geographically and potentially would make the policy more affordable than other options. However, this option is not without its disadvantages. First, it is unclear how private markets would be encouraged to underwrite all risks. Second, a mandatory all-perils policy might not be a cost- effective solution for the federal government, because it could create affordability concerns for low-income residents in certain areas and might require targeted government subsidies.[Footnote 64] If they did not sufficiently reduce postevent disaster relief, these subsidies could increase costs to taxpayers. Third, an all-perils policy would undoubtedly be more expensive than current homeowner policy premiums in some regions of the country. As a result, at least during the transition, it could lead to complaints about higher premium costs from residents of catastrophe-prone areas. Moreover, homeowners in relatively low-risk areas could wind up subsidizing the costs of insurance for those living in high-risk areas. Fourth, enforcement would be extremely challenging, as we have seen with mandatory flood insurance in communities in designated floodplains. Finally, this policy option faces opposition from the private insurance industry, in part because of concerns about state insurance regulators impeding private insurers' ability to charge premiums that reflect the actual risk of loss in catastrophe-prone areas. Private insurers have also traditionally opposed all-perils policies because of the difficulty of pricing flood and earthquake coverage. One insurance company has said that an all-perils policy would cause rates to skyrocket and could cause many insurers to abandon the homeowners insurance market. NAIC officials told us that the homeowners market was a $55 billion market- -not counting flood and earthquake exposure--and that most insurers were unlikely to walk away from a market this large. Federal Reinsurance Could Eliminate Timing Risk for Insurance Companies but Could Displace the Private Market: A federal reinsurance mechanism would provide an additional layer of insurance coverage for very large catastrophes, or megacatastrophes, and could be implemented in two ways.[Footnote 65] The first version of this option would create a federal mechanism that would serve as a backstop for state catastrophe funds to increase the amount of insurance and reinsurance available to states, expand the availability of catastrophe coverage, and possibly improve its affordability. [Footnote 66] States would create catastrophe funds and enter into agreements with the federal government--possibly, but not necessarily, the U.S. Treasury--and pay premiums for the reinsurance that would be used to support the reinsurance fund. Each state's payments would be based on risk and determined using actuarial and catastrophe modeling, and the states would be responsible for collecting premiums from insured commercial and residential property owners. The federal fund would provide payments to state funds for storms of a certain magnitude up to some predetermined level of payments. If the federal reinsurance fund was not adequately financed at the time of a catastrophe, it would issue government-backed bonds. A related but different version of this federal reinsurance option would authorize the Secretary of the Treasury to create an auction process for the sale of reinsurance contracts to private and state insurers and reinsurers.[Footnote 67] The secretary would make available reinsurance contracts covering both earthquakes and wind events. The auction process would be open to state and private insurers and reinsurers and would take place in at least six separate geographic regions, so that risks would be based on local factors and insurers in less risk-prone areas would not be subsidizing those in riskier areas. State programs would have to reach a minimum loss level before they would be eligible for federal funds. This version also establishes a disaster reinsurance fund within the U.S. Treasury to be credited with, among other sources of funds, amounts received from the sale of reinsurance contracts. The Treasury would be authorized to issue debt if the fund's resources were insufficient to pay claims--and reinsurance premiums paid to Treasury would be used to make interest payments to debt holders--but the fund would not receive federal appropriations. A national commission on catastrophe risks and insurance loss costs would advise the secretary. Both versions of this option offer advantages and disadvantages. First, federal reinsurance is advantageous because it has the potential to help insurance companies by limiting timing risk--the possibility that events will occur before insurers have collected enough premiums to cover them--potentially making insurers more willing to underwrite natural catastrophe insurance policies.[Footnote 68] Second, primary insurance companies may be less interested in canceling catastrophe insurance policies in coastal regions after a disaster if stable sources of reinsurance are available from state catastrophe funds. This option could also encourage the provision of catastrophe insurance via private insurance markets by limiting private insurers' liability for very large events and thus increasing their willingness to offer insurance for less catastrophic events. And a greater supply of natural catastrophe insurance could reduce the cost of insurance as competition for business intensified. Third, this option may also be advantageous because, if it were appropriately structured--that is, if program losses were funded by upfront premium payments--federal reinsurance should not require the use of taxpayer dollars. Finally, to the extent that this option increased the availability and affordability of catastrophe insurance, it would be preferable to postdisaster assistance and could limit the need for some types of postevent government payouts. While federal reinsurance has some appealing options, it is not without disadvantages. For example, neither version of the reinsurance option is intended to displace or compete with the private reinsurance market, because reinsurance contracts would not be sponsored in markets where private reinsurance markets offered coverage. However, federal reinsurance could compete with and possibly displace private reinsurance if the government offered coverage at levels that were well within private market capacity or set premium rates below what the private sector would charge for comparable risk. While the stated intent of this option is to charge a premium that fully reflects the risk assumed by the federal reinsurance fund, political and consumer pressures could be put on the federal fund to underprice premiums in terms of risk to keep premiums low for policyholders in high-risk areas. Charging a reinsurance premium that was not fully risk-based would expose the federal fund and the government to potentially significant unfunded contingent insurance risk. As a result, federal reinsurance could disproportionately benefit those living in high-risk areas. Should the fund experience losses that exceeded the premiums collected, the difference would have to be paid by the taxpayers, creating a cross-subsidy that favored those in catastrophe-prone areas. Also, the existence of federal reinsurance might affect market discipline, leading private insurers and state catastrophe insurance funds to loosen underwriting guidelines--that is, to insure properties that would not have been insurable without the availability of (low- cost) federal reinsurance. Such a change could be costly for the reinsuring federal facility. As a result, a federal reinsurance role could inadvertently encourage further development and population growth in areas with high natural catastrophe risk. Finally, government natural catastrophe insurance programs are not purely insurance programs and may have social goals. But if the government plans to intervene in the catastrophe insurance market, it may want to use mechanisms that mimic as closely as possible what operating private markets could have been expected to do. When federal insurance programs mimic private insurance, and base decisions on risk (as consistent with social goals), then government losses are more likely to be contained. A Federal Lending Facility Would Eliminate Timing Risk for State Catastrophe Insurance Programs but Would Face the Risk That the Loan Might Not Be Repaid: A federal lending facility would allow the federal government to use its borrowing power to extend temporary loans to state catastrophe funds. State catastrophe funds may not have the creditworthiness to borrow at acceptable interest rates. One proponent of this plan has suggested that the private insurance market could handle all or nearly all catastrophe exposure, but possibly not at the moment the catastrophe happened. Creating a lending facility in the federal government would allow the government to provide the capital to meet the temporary shortage and spread the repayment over time without assuming the underwriting risk held by the insurers. Under this option, state catastrophe funds would be required to secure private reinsurance and would have the ability to sell catastrophe bonds to repay the money loaned to them by the federal government. The loans would be made at market prices to guarantee that capital was efficiently allocated and- -given that an insurance company that has just paid out a large claim does not have the same quantity or quality of assets as a solvent insurer or bank--would be secured both by the future income stream of premium payments from state residents through insurance companies to the state catastrophe funds and by bond proceeds. The loans would be of short duration, perhaps 2 to 3 years at maximum, and would provide state catastrophe funds with encouragement and time to access the private capital market. State catastrophe funds would be expected to demonstrate to the federal lending facility that the states were doing all that they could to attract private capital. A proposed trigger for the federal lending facility would be a megacatastrophe. The creation of a federal lending facility would have several advantages. First, a federal lending facility would shift timing risk, which is significant in the catastrophe insurance business, from the insurance industry to the federal government. The federal government, because of its borrowing power, is uniquely able to deal with timing risk. Second, a federal lending facility could mean that taxpayers would assume little or possibly no insurance risk, because the insurers would be responsible for paying all of the losses from catastrophic events, although not necessarily in the year of the catastrophe. Finally, through the requirement that the states do all that they can to attract private capital, the option may lead to insurance regulatory reforms in areas such as rate regulation that have inhibited the influx of private capital. A federal lending facility would also have a number of disadvantages. First, it is not clear how this federal lending facility would encourage premiums that reflected risks, would foster broad citizen participation, or would be a cost-effective solution. Second, it would expose the facility and ultimately taxpayers to credit risk if a state did not repay its debt. Third, a federal lending facility could also require the creation of a new federal entity or structure to administer the system. Fourth, like the federal reinsurance option, such a lending facility could have a competitive advantage over the private reinsurance sector, particularly if the terms were too easy or if borrowed funds did not have to be repaid. States in high-risk regions would have a financial incentive to seek nonmarket terms and conditions in loans. Finally, this option would decrease the incentives for insurers and reinsurers to accurately assess, underwrite, and price risk. Tax-Deferred Reserves for Insurance Companies Could Encourage Greater Private Sector Coverage but Could Be Costly for the Federal Government and Have Other Disadvantages: A fourth policy option would be to permit private insurers to establish tax-deferred reserves for future catastrophes.[Footnote 69] This option could encourage some insurers to maintain or expand their catastrophe insurance coverage in regions with significant or projected catastrophe exposures. This option is also intended to provide insurers with an incentive to write catastrophe coverage in hazard-prone areas while improving their own financial strength. It would require amending the U.S. Tax Code, because current tax laws and accounting principles discourage U.S. property and casualty insurers from accumulating long- term assets specifically for payment of future losses by taxing these assets.[Footnote 70] Because the size and timing of disasters that have not taken place is uncertain, assets set aside for catastrophe losses, together with any interest accrued, are taxed as corporate income in the year in which they are set aside. Although there is a federal income tax deduction for losses that have already occurred, reserves for uncertain future losses are not tax deductible. Tax-deferred reserving has its advantages. Tax-deferred reserving could mean that state regulators would be more willing to approve risk-based rates, because premiums could now be set aside rather than flow into profits. Consistent with the intended purpose of this option, tax- deferred reserving could increase the willingness of insurance companies to increase capacity without risking insolvency, because the companies would be less dependent on the uncertain prices available in reinsurance markets. In this case, the option would encourage a solution by private insurance markets and more broad-based participation in catastrophe insurance programs. Finally, this approach could reduce the need for state catastrophe insurance mechanisms by increasing the willingness of private insurers to remain or enter certain catastrophe-prone markets, such as Florida and other Gulf Coast states. However, tax-deferred reserving also raises a number of broader issues that must be considered. Tax-deferred reserving would reduce current federal tax revenue. However, as with other options, the net cost would have to be determined by weighing the tax cost against potential savings from federal postdisaster assistance programs. Deferring taxes on reserves for insurance companies could also be disadvantageous if this system created tax benefits that favored one type of activity over another. For example, to the extent that tax-deferred reserving became prevalent, it could displace the reinsurance market or other forms of hedging. Finally, such reserves could also be subject to manipulation or abuse if insurers used them to obscure current income by smoothing income flows across years. Homeowner Catastrophe Savings Accounts Could Broaden Participation in Catastrophe Insurance Programs but Could Reduce Federal Tax Revenue: Like tax-deferred reserves, the fifth policy option would also require amending the U.S. Tax Code to provide a tax incentive, but this one would be aimed at homeowners, who would be allowed to accumulate before- tax funds to pay expenses related to disasters. The accounts would operate much like those currently in use for health care expenses, allowing homeowners to withdraw both savings and interest for qualified disaster expenses such as deductibles, uninsured losses, flood damage, and structural upgrades to mitigate damage from future storms. A bank or another designated organization would be the custodian for these accounts. Under one current option, homeowner contributions would be limited to (1) $2,000 for individuals with homeowners insurance and deductibles of not more than $1,000, and (2) the lesser of $15,000 or twice the insurance deductible for homeowner insurance deductibles of more than $1,000. In June 2007, the South Carolina Legislature passed legislation authorizing the creation of catastrophe savings accounts for use by state residents in paying natural catastrophe insurance deductibles.[Footnote 71] This option could induce more homeowners to participate in natural catastrophe insurance programs. Moreover, allowing homeowners to use tax-deferred savings to cover mitigation expenses might encourage more mitigation activities to reduce natural catastrophe risk. However, implementation challenges pose disadvantages that would have to be addressed. For example, it is unclear to what extent such a mechanism would encourage those who are not insured to purchase insurance. Rather than increasing participation, it could result in a tax benefit for those who are already insured. Like the tax-free reserves option, these savings accounts would also cost the federal government in reduced tax revenues. But once again, the actual net cost to the government would depend on the potential offsetting savings from postcatastrophe funding mechanisms. Favorable Tax Treatment for Catastrophe Bonds Could Increase Insurers' Access to Capital Markets, but Some Question the Need for Such Tax Treatment: The sixth policy option would create certain tax advantages for catastrophe bonds.[Footnote 72] Historically, catastrophe bonds have been created in offshore jurisdictions where they are not subject to any income or any other tax (i.e., in tax havens). This option would facilitate the creation of onshore transactions, potentially reducing transactions costs and allowing for increased regulatory oversight. Tax treatment of catastrophe bonds would be similar to the treatment received by issuers of asset-backed or mortgage-backed securities that, for example, are generally not subject to tax on the income from underlying assets, which is passed on to investors.[Footnote 73] More favorable tax treatment of catastrophe bonds would increase the ability of insurance markets to access capital markets by making these products more attractive to investors. Making catastrophe bonds more attractive to issuers and investors could, in turn, make insurance and reinsurance companies more willing to underwrite catastrophe risk and increase the availability of coverage, because these companies could pass on more catastrophe risk to investors. One disadvantage of this option is that it is not clear how its implementation would encourage premiums that fully reflect risk or how it would encourage broad-based participation in catastrophe insurance markets. It is also not clear how this option would be a cost-effective solution for the federal government when both predisaster and postdisaster costs are counted. Some reinsurers have pointed out that favorable tax treatment of catastrophe bonds could be disadvantageous because it could create a new class of reinsurer that would operate under regulatory and tax advantages not afforded U.S. reinsurance companies. Finally, recent catastrophe bond issuances by the two largest U.S. primary insurance companies may indicate that catastrophe bonds do not need a different tax treatment to make them economically viable.[Footnote 74] However, if market transparency and the development of uniform terms and conditions do not take place, only the largest insurers may be able to take advantage of catastrophe bonds. Property-Tax Assessment for Federal Reinsurance May Broaden Participation but May Be Costly for the Federal Government: The final policy option we examined was a state plan, funded by state property taxes, that would require mandatory all-perils natural catastrophe insurance coverage on residential property. All primary residential properties in a state would be required to have catastrophe insurance coverage. Participating insurers would assume the primary risk on the property and would have reinsurance from a qualifying reinsurance company. The state would pay an annual natural catastrophe insurance premium financed by an annual property tax assessment on all residential and commercial properties in the state, and homeowners could deduct the cost from their federal taxes.[Footnote 75] The insurance coverage would be provided by private insurance companies selected by a government administrator who would qualify them as providers of catastrophe insurance. To ensure that premiums were reasonable, the primary and reinsurance coverage would require large deductibles that would be paid in layers by the homeowner, the state, and the federal government. Homeowners would be responsible for the first 10 percent of the value of the home, with a state catastrophe fund paying the next layer of the deductible. The state would provide a fixed-dollar deductible--for example, $100 million--for all homeowners, with the federal government as the backstop provider, paying a deductible that was a multiple of the amount that the state put up. Proponents of this plan point out that it is market-based, designed to involve the private sector, and if risk-based premiums are required is not a "government relief program." Plan supporters also point out that the option protects the tax base of a state's economy as well as the creditworthiness of a state's bond rating. One possible advantage of this policy option for the consumer is that the premiums paid from property taxes are intended to be tax deductible.[Footnote 76] Moreover, paying the premium from property taxes could increase participation at the state level and create a broad-based program that would limit adverse selection and moral hazard. Finally, maintaining higher deductibles could result in lower insurance premiums. However, this plan also has its disadvantages. Paying the premium from homeowner property taxes collected by the state would reduce federal tax revenues, and, if a disaster occurred, the federal government would have to pay some portion of the deductible. Like the other tax-related options, this option could reduce federal tax revenue if the new deduction were not offset by savings from the elimination of preevent premium subsidies or postevent disaster relief. As a result, it is not clear whether this option may or may not be the most cost-effective for the federal government. Also, using property taxes to pay insurance premiums might diminish the effectiveness of using the price of insurance as a signal of the risk of living in a particular location. One critic has argued that allowing homeowners to deduct the premium portion of the property taxes combined with the federal deductible could result in a double federal subsidy. Finally, this policy option would raise homeowners property taxes, potentially creating homeowner resistance to the assessment. Agency Comments and Our Evaluation: We provided a draft of this report to NAIC for comment and provided excerpts from the draft to Alabama Beach Pool, the CEA, FCIC, FHCF, Florida Citizens, FHCF, the GUA, HUD, Louisiana Citizens, Mississippi Windpool, the North Carolina Beach Plan, SBA, the South Carolina Windpool, and the Texas Windpool. NAIC provided written comments that are reprinted in appendix III. In these comments, NAIC officials said that our draft report was thorough, and that they were pleased that we outlined the advantages and disadvantages of several proposals rather than favoring a single outcome. NAIC officials suggested that we also include in this report two recently proposed options, including one that includes an allocation system for determining what portion of hurricane damages should be attributed to wind and what portion to flooding and the creation of a federal entity to oversee property insurance rates in the coastal zone. While there are interesting features to both options, they were too recent to be included in our review and analysis. However, we will explore both options during the course of our ongoing work involving NFIP. NAIC officials also commented on the language in the draft report discussing allegations made by some critics of state rate regulation who suggest that state regulators may be suppressing rates for some catastrophe insurers. As these officials pointed out, the allegations in this report are attributed to others and are not presented as our position. We recognize the challenges involved in ensuring that consumers are charged appropriate premiums that reflect their risk of exposure to natural catastrophes. Given that premium rates requested are based on a variety of factors that involve a certain amount of judgment--including anticipated losses on claims and related expenses; the need to build a surplus; and other factors, including profit--the rate-setting process is open to interpretation and some amount of negotiation. That is, reasonable but different assumptions about the probability of future losses can result in substantial disagreements about rates. However, if state regulators and the insurance markets consistently have divergent opinions about the cost of the risk exposures, the implications can be far-reaching. As we discuss in this report, for state natural catastrophe insurance programs, if premium rates determined by state insurance regulators consistently result in financial resources that are inadequate to pay policyholder claims after a disaster, postfunding mechanisms must be used to pay shortfalls. Postfunding can result in costs to the private insurance market and may mean that taxpayers in low-risk areas are subsidizing the costs of those living in high-risk areas. Similarly, a pattern of regulator-approved rates for private insurance companies that are consistently below what the market believes to be the true risk rate may result in the withdrawal of healthy, diversified insurance companies from the market. However, if premium rates are set at a level reflecting the market's perception of the true risk rate, more competitors are likely to enter. Alabama Beach Pool, the CEA, FCIC, FEMA, Florida Citizens, FHCF, the GUA, Louisiana Citizens, Mississippi Windpool, the North Carolina Beach Plan, SBA, the South Carolina Windpool, and the Texas Windpool provided technical comments that we incorporated in this report as appropriate. As agreed with your office, unless you publicly announce its contents earlier, we plan no further distribution of the report until 30 days from the date of this letter. At that time, we will provide copies to interested congressional committees; the Chairman and Ranking Member of the Senate Committee on Banking, Housing, and Urban Affairs; and the Chairman of the House Committee on Financial Services. We will also make copies available to others upon request. In addition, this report will be available at no charge on the GAO Web site at [hyperlink, http://www.gao.gov]. Please contact me at (202) 512-8678 or williamso@gao.gov if you or your staff have any questions concerning this report. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. Key contributors to this report are listed in appendix IV. Sincerely yours, Signed by: Orice M. Williams: Director, Financial Markets and Community Investment: [End of section] Appendix I: Objectives, Scope, and Methodology: Our objectives in this report were to examine (1) the rationale and funding of the federal and state programs that have supplemented, or substituted for, private natural catastrophe insurance; (2) the extent to which Americans living in areas of the United States that are at high risk for natural catastrophes are uninsured and underinsured, and the types and amounts of federal payments to such individuals since Hurricanes Katrina, Rita, and Wilma; and (3) public policy options for revising the federal role in natural catastrophe insurance markets. We reviewed or analyzed documents on federal and state natural catastrophe insurance programs, the numbers of uninsured and underinsured and federal payments to them, options to redefine the federal role in natural catastrophe insurance, and principles on which change options can be based and evaluated. We interviewed officials from public interest groups, insurance companies, reinsurance companies, insurance brokers, insurance and reinsurance associations, insurance agents and their associations, state catastrophe insurance plans, state insurance departments, federal catastrophe insurance agencies, the Department of Housing and Urban Development (HUD), the Small Business Administration (SBA), Fannie Mae, Freddie Mac, rating agencies, a risk modeling organization, academia, law firms, a hedge fund, a private research organization, consumer groups, and others. To determine the mechanisms governments use to supplement or substitute for private catastrophe insurance markets, we collected oral and documentary information from public and private officials in various states with high and low catastrophe risk and in Washington, D.C. We sourced financial data for government natural catastrophe insurance programs from financial statements, bond offering documents, and other similar financial documents. To determine the number of uninsured and underinsured Americans and payments made to such individuals after the 2005 hurricanes, we collected information from states, examined federal agency data, interviewed federal officials who prepared these data, sought information from the private sector, and interviewed state officials responsible for disbursing federal disaster funds. We focused our analysis on the federal disaster assistance to homeowners and renters who lacked adequate insurance in the five Gulf Coast states directly impacted by Hurricanes Katrina, Rita, and Wilma. These five states are Alabama, Florida, Louisiana, Mississippi, and Texas. Data on the numbers and amounts of money disbursed to the uninsured and underinsured were incomplete and had a number of limitations. For instance, because we often could not separate payments to homeowners versus payments to renters, we generally included the entire amount in our analysis. Also, we generally excluded administrative and other expenses that federal disaster assistance programs incur in distributing assistance. Our analysis was limited to the major federal disaster assistance programs that we identified as providing relief to homeowners and renters. These programs are the Federal Emergency Management Agency's (FEMA) Individuals and Households Program (IHP), SBA's Disaster Loan Program (DLP), and HUD's Community Development Block Grant (CDBG) program. Our identification of relevant federal disaster assistance programs may be incomplete. Other federal agencies are involved in federal disaster assistance according to the mission assignment issued and approved by FEMA, as we reported separately in Disaster Relief: Governmentwide Framework Needed to Collect and Consolidate Information to Report on Billions in Federal Funding for the 2005 Gulf Coast Hurricanes, GAO-06-834 (Washington, D.C.: Sept. 6, 2006). To determine the amount of federal disaster assistance appropriated by Congress to FEMA and the amount paid to homeowners and renters who lacked adequate insurance through FEMA IHP, we obtained and analyzed data provided by FEMA officials describing the funds obligated for the subcategories of Housing Assistance, Other Needs Assistance, and Manufactured Housing in Alabama, Florida, Louisiana, Mississippi, and Texas following Hurricanes Katrina, Rita, and Wilma. In analyzing these data, we had to make certain judgments in deciding which specific subcategories of funds to include in our analysis. In particular, FEMA noted that the Other Needs Assistance data contained funds for services that would not be provided by personal property coverage in standard private homeowners insurance, such as medical and funeral expenses. However, we included Other Needs Assistance data in our analysis because these are expenses that may have been covered by other types of insurance, such as health and life, and, therefore, still provide a reasonable approximation of insurance coverage. Also, FEMA officials noted that the Manufactured Housing data included expenses that would not be included in additional living expenses coverage provided by standard private homeowners insurance. For example, other expenses included unit purchase, haul/install, utilities, site lease, maintenance, deactivation, and the transition out of service. We included these data in our analysis because they are designed to serve a similar purpose as the additional living expenses coverage provided by insurance companies. We assessed the reliability of the data provided by agency officials by interviewing agency officials knowledgeable about the data systems; obtaining oral responses from the agency; and reviewing agency reports regarding (1) the agency's methods of data collection and quality control reviews, (2) practices and controls over data entry accuracy, and (3) any limitations of the data. It is possible that FEMA's data analysis methodology is different from that employed by the other agencies we reviewed. Nevertheless, we determined that these data were sufficiently reliable for the purposes of our engagement. Finally, we interviewed officials from FEMA Disaster Assistance Directorate, which administers IHP, and reviewed the document entitled Oversight of Gulf Coast Hurricane Recovery, A Semiannual Report to Congress, October 1, 2006-March 31, 2007, by the President's Council on Integrity and Efficiency and the Executive Council on Integrity and Efficiency. To determine the amount of federal disaster assistance appropriated by Congress to SBA and the amount paid to homeowners and renters who lacked adequate insurance through SBA DLP, we reviewed the previously mentioned document entitled Oversight of Gulf Coast Hurricane Recovery, and interviewed agency officials. We obtained and analyzed data provided by SBA that included, among other things, the amount of loan funds approved net of other federal disaster assistance and insurance proceeds to loan recipients. We multiplied this total by the subsidy rate of the loans--14.64 percent in 2006. That is, for every $100 that SBA lends, the cost to the federal government is $14.64. The subsidy rate is roughly the percentage of loan principal that is not repaid as well as the difference between the market interest rate and the rate charged by SBA. We believe that subsidy cost is the most accurate representation of the amounts made available and paid to homeowners and renters because the loans under DLP must be repaid by recipients at a subsidized interest rate. We assessed the reliability of the data provided by agency officials by interviewing agency officials knowledgeable about the data systems and obtaining from the agency written responses regarding (1) the agency's methods of data collection and quality control reviews, (2) practices and controls over data entry accuracy, and (3) any limitations of the data. It is possible that SBA's data analysis methodology is inconsistent with that employed by the other agencies we reviewed. Nevertheless, we determined that these data were sufficiently reliable for the purposes of our engagement. To determine the amount of federal disaster assistance appropriated and paid to homeowners and renters who lacked adequate insurance through the HUD CDBG program, we interviewed agency officials and reviewed the previously mentioned document entitled Oversight of Gulf Coast Hurricane Recovery. We obtained publicly available data from HUD and each of the five Gulf Coast states that received emergency CDBG appropriations. We reviewed GAO testimony on Gulf Coast rebuilding that described the CDBG programs established in the Gulf Coast states.[Footnote 77] Congress approved emergency appropriations for HUD CDBG in two installments: $11.5 billion in December 2005 and $5.2 billion in June 2006, for a total appropriation of $16.7 billion. Our goal was to determine what portion of the total appropriation was intended for homeowners in the five Gulf States. We made certain judgments in deciding whether particular subcategories of funds applied to our calculations for each state. It is possible that we did not identify all of the relevant funds. For Florida, we used the Florida Department of Community Affairs, 2005 Disaster Recovery Initiative Action Plan (Apr. 14, 2006) and 2006 Disaster Program Action Plan (Dec. 19, 2006). HUD designated for Florida $82.9 million of the original $11.5 billion included in the December 2005 emergency appropriation. Florida's action plan calls for the funds to be distributed through entitlement communities, nonentitlement communities, and federally recognized Indian tribes. Grant recipients are required to use at least 70 percent of the funds for the provision of affordable housing. Therefore, approximately $58 million of the Florida CDBG grants will be allocated to the provision of affordable housing. In addition, the June 2006 emergency appropriation included $5.2 billion to the CDBG program, and, on August 18, 2006, HUD made $100,066,518 available to Florida for repair, rehabilitation, and reconstruction of affordable rental housing, and for the unmet needs of evacuees who were forced from their homes and are now living in other states. The entire amount has been made available for mitigation programs through the My Safe Florida Home Program and other programs. For Alabama, we interviewed officials from the Alabama Department of Economic and Community Affairs (DECA). We obtained and analyzed information from DECA officials regarding the plan for distribution of HUD CDBG disaster recovery funds. We learned that DECA determined to make $14,460,588 available for unmet housing needs. In addition, on August 18, 2006, HUD made $21,225,574 available to Alabama for repair, rehabilitation, and reconstruction of affordable rental housing, and for the unmet needs of evacuees who were forced from their homes and are now living in other states. Of this amount, $16,964,296 has been made available for Disaster Relief, Recovery and Restoration of Housing and Infrastructure, and Affordable Rental Housing. For Mississippi, we used the Mississippi Development Authority, Homeowner Assistance Program Partial Action Plan (Mar. 31, 2006). Mississippi's partial action plan made $3 billion available for the Homeowner Grant Assistance Program, which is for people who owned homes located outside of the fede