

	June 21, 2007 HP-470 Testimony of Treasury Assistant Secretary for Financial Institutions David G. Nason Before the U.S. House Committee on Financial Services Subcommittee on Capital Markets, Insurance and Government Sponsored Enterprises Washington - Thank you, Chairman Kanjorski, Ranking Member Pryce, and other members of the Subcommittee for inviting me to appear before you today. The market for terrorism risk insurance in the United States was significantly changed by the terrorist attacks of September 11, 2001. Of course, prior to September 11, terrorism risk clearly existed in the United States. We experienced the 1993 bombing of the World Trade Center, the 1995 Oklahoma City bombing, the 1996 Centennial Olympic Park bombing, and the "Millennium Bomber's" December 1999 attempted bombing of the Los Angeles International Airport. Despite these events, most commercial property and casualty insurance companies continued to provide coverage for terrorism risk in the insurance policies sold to their commercial policyholders. The terrorist attacks of September 11 resulted in insured losses of approximately $32 billion, which at the time was the largest single insured loss event in U.S. history. The recognition that terrorist attacks could cause losses of such scale spread across multiple insurance products and concentrated in a relatively small geographic area, caused the insurance industry to undertake a broad reassessment of the likelihood and potential losses associated with terrorism. Immediately following September 11, commercial property and casualty insurers sought to exclude coverage for terrorism risk in many policies. Reinsurance contracts also began excluding coverage for terrorism. In the months after September 11, there were increasing concerns about potential economic disruptions caused by the unwillingness of many insurance companies to provide terrorism insurance. In response, Congress passed and the President signed the Terrorism Risk Insurance Act (TRIA) in late 2002. TRIA established a temporary federal program of shared public and private compensation for privately-insured commercial property and casualty losses resulting from acts of terrorism. The TRIA legislation stated that the purposes of the legislation were to address market disruptions, to ensure the continued widespread availability and affordability of commercial property and casualty insurance for terrorism risk, and to allow for a transition period for the private markets to stabilize and build capacity while preserving State insurance regulation and consumer protections. While TRIA was largely successful in achieving its original purposes, given some remaining uncertainty surrounding the development in the market for terrorism risk insurance, TRIA was temporarily extended in 2005 for an additional two years by the Terrorism Risk Insurance Extension Act of 2005 (the "Extension Act"). Today, I would like to provide an overview of the key features of TRIA and the Extension Act, the key findings of the President's Working Group on Financial Markets' (PWG) 2006 report to Congress on terrorism risk insurance, and some principles for the federal government's role in the market for terrorism risk insurance going-forward. Our view of TRIA is shaped by the belief that the most efficient, lowest cost, and most innovative methods of providing terrorism risk insurance will come from the private sector. The Administration believes that three elements are critical if TRIA is to be reauthorized for a second time: the program remains temporary and short-term; private sector retentions are increased; and there is no expansion of the program. Treasury cannot support efforts that move the program in a direction that is inconsistent with these key elements. The Terrorism Risk Insurance Act and the Extension Act TRIA essentially established a government reinsurance program. Much like typical provisions found in reinsurance, the TRIA program requires that insurers first retain a portion of terrorism risk exposure themselves (referred to as "deductible") with the insurer and government then sharing in the losses above the initial retention (referred to as "co-share" or "co-pay"). Unlike a typical reinsurance policy, however, there is no up-front premium charged for the reinsurance coverage provided under TRIA; but instead, any federal expenditure can be collected, or recouped, after a loss through surcharges applied to premiums paid by commercial policyholders regardless of whether their insurers had received TRIA payments. Some key features of TRIA and the Extension Act include the following: Private Sector Retentions An insurer's "retention" under TRIA generally refers to the amount of terrorism risk exposure that an insurer retains. An insurer's retention is comprised of its insurer deductible, its co-share of the insured losses above its deductible, or all of its losses if an attack results in industry-wide losses below an event size threshold, called the "Program Trigger." An insurer's retention under TRIA is a key provision that governs the amount of terrorism risk exposure held by the private sector. The general structure of TRIA and the Extension Act requires increases in private sector retentions over time to encourage development of private market capacity to provide terrorism risk insurance over time. An insurer's deductible is company specific and is calculated based on the size of the insurer's prior year's premium revenue from the types of insurance covered by TRIA. Insurer deductibles have increased throughout the TRIA program – from its 2003 level of 7 percent of an insurer's prior year's direct earned premiums, to 10 percent in 2004, and 15 percent in 2005. The Extension Act further increased insurer deductibles to 17.5 percent in 2006 and 20 percent in 2007. In the event of a certified terrorist act, each insurer will cover 100 percent of the insured losses up to its deductible before being eligible for federal payments under TRIA. Tying an insurer's deductible to its revenue helps ensure that the amount of insured losses the company itself is responsible for is commensurate with its size and assets. Some of the largest insurers that participate in the program have deductibles in the billions of dollars. Once an insurer pays insured losses up to its deductible, insured losses above its deductible amount would then be shared between the insurer and the federal government. The federal share of insured losses above the insurer's deductible had been 90 percent through the first four years of the TRIA program, and was reduced to 85 percent in 2007 – thus increasing the private sector's share from 10 percent to 15 percent. This provision of TRIA encourages proper claims adjustment as insurers will have "skin in the game" in deciding and settling insurance claims, much the same as provisions included in private sector reinsurance contracts. In addition to the deductible and co-share, an insurer retains all of its losses if industry-wide aggregate losses are below the minimum event size eligible for payments under TRIA, or what has come to be known as the "Program Trigger." Under TRIA, in order for an event to be certified as an act of terrorism, the losses suffered by the insurance industry as a whole must exceed at least $5 million in the aggregate. As originally structured, certified acts of terrorism resulting in losses above $5 million would have been eligible for federal payments under TRIA, essentially making the certification and the event's eligibility for federal payments under TRIA equivalent. The minimum event size qualifying an event for federal payments under TRIA was raised beginning in 2006 by the Extension Act, which specifically added the concept of a Program Trigger to TRIA. Under the Program Trigger concept, the Treasury Secretary is directed not to compensate insurers under TRIA unless the aggregate industry insured losses exceed certain "trigger" amounts: $50 million in 2006 and $100 million in 2007. Once the threshold is met, program payments can then be made to an insurer once it has paid claims and met its company-specific deductible. Lines of Coverage Insurance coverage under TRIA is limited to commercial property and casualty insurance, which was the primary area of concern in terms of dislocations associated with the September 11 terrorist attacks. TRIA does not apply to personal insurance, such as homeowners, automobile, or life insurance. While TRIA did not specifically define commercial property and casualty insurance, it did specifically include excess insurance, workers' compensation insurance, and during the first three years of the TRIA program, surety insurance. In addition, TRIA specifically excluded certain types of insurance: · Federal or private crop insurance; · Private mortgage insurance, or title insurance; · Financial guaranty insurance offered by a monoline financial guaranty insurance corporation; · Insurance for medical malpractice; · Health or life insurance, including group life insurance; · Federal flood insurance; and, · Reinsurance or retrocessional reinsurance. In implementing the definition of commercial property and casualty insurance, Treasury relied on the lines of business ("lines") under which insurers report their premiums in annual statement filings pursuant to forms and rules adopted by the National Association of Insurance Commissioners (NAIC). The specific lines that were included in the program were established by Treasury through regulation, in consultation with the NAIC. With respect to implementing the program, policies whose premiums are reported to the NAIC on designated commercial lines qualify for TRIA coverage. The Extension Act scaled back the scope of the program so that TRIA no longer covers: · Commercial automobile insurance; · Burglary and theft insurance; · Surety insurance; · Professional liability insurance (but not directors' and officers' liability insurance); and, · Farmowners' multiple peril insurance. Terrorism risk insurance for these lines of insurance, which was covered only during the first three years of the program, was successfully transitioned back to the private market without any signs of market disruption. Certified TRIA Events The TRIA program covers losses from certified acts of terrorism. In order to qualify as an act of terrorism, an event must be certified by the Secretary of the Treasury with the concurrence of the Secretary of State and Attorney General of the United States as being: · a violent act, or an act dangerous to life, property or infrastructure; · resulting in damage within the U.S., or to a U.S. air carrier or U.S. flag vessel, or on the premises of a U.S. mission; and, · committed by an individual or individuals acting on behalf of any foreign person or foreign interest, as part of an effort to coerce the civilian population of the U.S. or to influence the policy or affect the conduct of the U.S. government by coercion. Terrorism coverage is often described as "certified acts" coverage (based on the TRIA definition) and "non-certified acts" coverage (acts of terrorism that are not certified under TRIA because they do not meet one or several of the certification requirements). "Certified acts" are synonymous with foreign acts of terrorism due to the requirement that the act be "committed by an individual or individuals acting on behalf of any foreign person or foreign interest." Under TRIA, an act committed by a "home-grown" terrorist could currently be certified as an act of terrorism and covered by TRIA so long as the terrorist was acting on behalf of any foreign person or foreign interest, and the other requirements are met. However, purely domestic terrorism, such as eco-terrorist attacks or an attack like Oklahoma City, would not be covered by TRIA. Such non-certified risks generally continue to be insured by the private market. Recoupment Unlike a private sector reinsurance company, the TRIA program does not require insurers to pay up-front premiums and does not build up surplus to pay future claims. Instead, the program is funded on a post-loss basis. TRIA provides authority for Treasury to recoup its federal payments through annual surcharges on commercial policyholders of up to three percent of a policy's premium. Certain recoupment is mandatory, while in other circumstances TRIA authorizes discretionary recoupment. Mandatory recoupment is based on the concept of an "insurance marketplace aggregate retention" amount, which specifies the amount of losses the private sector as a whole must absorb in any given year. If the insured losses that the insurers collectively retain (individual company deductibles plus the co-pay portions paid above deductibles) are lower than the marketplace aggregate retention, Treasury must recoup the difference. In addition, Treasury has the discretion to seek recoupment of up to the full amount paid out based on consideration of specific factors described in TRIA. The "insurance marketplace aggregate retention" amounts have increased each year of the program, going from $10 billion in the first year to $27.5 billion in 2007. Key Outcomes of the Extension Act The TRIA program was originally designed as a three-year program set to expire on December 31, 2005. The temporary program structure allowed the federal government to re-evaluate the program in the context of current market conditions. As the debate surrounding the extension of TRIA took place in 2005, the Administration focused on encouraging the private insurance market to develop innovative solutions and build capacity. This serves to reduce potential exposure to taxpayers. The core changes ultimately adopted as part of the Extension Act – increasing deductibles and co-share amounts, elevating program trigger levels, and eliminating coverage for certain lines of insurance – all focused on encouraging greater private market participation over time. The impact of these changes and an overall evaluation of the market for terrorism risk insurance formed the basis for much of the President's Working Group on Financial Market's (PWG) 2006 report on terrorism risk insurance. The Findings of the President's Working Group on Financial Markets The Extension Act required the PWG to perform an analysis regarding the long-term availability and affordability of insurance for terrorism risk, including group life coverage; and coverage for chemical, nuclear, biological, and radiological events. In conducting this analysis, the PWG was assisted by staff of the member agencies who reviewed academic and industry studies on terrorism risk insurance, sought additional information and consultation through a Request for Comment published in the Federal Register, and also met with insurance regulators, policyholder groups, insurers, reinsurers, modelers, and other government agencies. The PWG submitted its report to Congress last September. Key findings of the report are summarized below. Long-Term Overall Availability and Affordability of Terrorism Risk Insurance One of the key findings of the PWG report was that overall the availability and affordability of terrorism risk insurance has improved since the terrorist attacks of September 11, 2001. The general trend observed in the market has been that as insurer retentions have increased under TRIA and policyholder surpluses have risen, prices for terrorism risk have fallen, and take-up rates have increased. Much of the improvement in the terrorism risk insurance market is due to several important factors, including better risk measurement and management, improved modeling of terrorism risk, greater reinsurance capacity, and a recovery in the financial health of property and casualty insurers. · Since September 11, insurers have made greater use of sophisticated models that allow them to identify and manage concentrations of risk in order to avoid accumulating too much risk in any given location. This improvement in risk accumulation management has allowed insurers to better diversify and control their terrorism risk exposures, which has enhanced their ability to underwrite terrorism risk. In addition, a significant effort has been made by the insurance industry in modeling the potential frequency and severity of terrorist attacks; however, given the uncertainty of terrorism in general and, in particular, the uncertainty associated with these modeling efforts, insurers appear to have limited confidence to date in these models for evaluating their risk exposures. · In terms of market capacity, the PWG found that the quantity of terrorism risk reinsurance capacity has increased since the period following September 11. In addition, the financial health of insurers has recovered since September 11. As a result, insurers have more available capacity to allocate to terrorism risk as demonstrated by the increased provision of terrorism risk insurance coverage over the past few years. Despite these overall improvements, the PWG report found that a significant number of policyholders are still not purchasing terrorism coverage – approximately 40 percent of all policyholders do not purchase coverage. Even in major cities, a high proportion of policyholders are not purchasing terrorism risk insurance. For example, in 2004, 46 percent of policyholders in New York City had not purchased terrorism insurance; in Los Angeles, 61 percent had not purchased terrorism insurance; in Chicago, 42 percent; and in Washington, D.C., 40 percent. Recently reported data for 2006 suggests this has improved for some cities; for example, approximately one-quarter of policyholders in the New York metropolitan area are uninsured, as compared to 46 percent in 2004. The PWG's report, Treasury's own 2005 study, and others have found that the primary reasons for non-purchase were price, perceptions of low risk, and perhaps to some degree an expectation that federal disaster aid might be available if a significant attack were to occur. The PWG report concluded that further improvements in insurers' ability to model and manage terrorism risk will likely contribute to the long-term development of the terrorism risk insurance market. However, the high level of uncertainty currently associated with predicting the frequency of terrorist attacks, along with what appears to be a general unwillingness of some insurance policyholders to purchase insurance coverage, makes any prediction of the potential degree of long-term development of the terrorism risk insurance market somewhat difficult. Group Life Insurance As passed by Congress in 2002, TRIA did not include group life insurance in the program. Treasury was required to evaluate market conditions and determine whether to include it in the program if both insurance and reinsurance were not available, or not likely to be available in the future. In 2003, Treasury found that group life insurance coverage was readily available for consumers. Thus, group life was not added to the program. In 2005, when TRIA was extended by Congress, group life was not added to the program. The PWG report found that group life insurance is still widely available in the private market even though it is not part of the TRIA program. In particular, the group life market is highly competitive and is very price sensitive. Group life insurers concede that competitive pressures have caused them to make coverage available, even in the absence of TRIA protection. In contrast to property and casualty insurers, group life insurers have decided to forgo purchasing reinsurance and to focus less on managing risk accumulations. Chemical, Nuclear, Biological, or Radiological (CNBR) Coverage under TRIA CNBR is currently covered under TRIA. However, TRIA does not require insurers to make CNBR terrorism coverage available to policyholders if CNBR coverage for non-terrorism events is similarly not provided. Although not required by TRIA, if CNBR terrorism coverage is provided by the insurance policy, such as with workers' compensation insurance, TRIA covers insured losses from a certified terrorist event involving CNBR. The PWG report found that historically CNBR risks (caused by a terrorist or by any other event) were typically not covered by insurance (except when mandated by state law, such as with workers' compensation). The factors determining the availability and affordability of CNBR coverage have more to do with the nature, scale, and uncertainty of the damage and losses from CNBR events – however caused – and less to do with terrorism specifically. In addition, policyholder expectations regarding their own potential terrorism risk exposure are probably lower and their expectations about the likelihood of post-disaster federal aid are probably higher for CNBR attacks than for relatively smaller-scale conventional terrorist attacks. The Federal Government's Role in the Market for Terrorism Risk Insurance As a basic principle, the federal government's role in any market, including the market for terrorism risk insurance, should be limited to those areas where private markets cannot function and hence broader costs are imposed on our Nation's overall economy. In playing such a role at a time when it was needed, TRIA appears to have been successful. TRIA provided time for insurers and others to adjust to the risks made clear by the September 11 terrorist attacks. Subsequently, there have been positive market responses by insurers and reinsurers to the reductions in the federal role over the five years that TRIA has been in place, most notably by assuming additional terrorism risk exposure in each year of the program. And as insurers have increased their terrorism risk exposure as TRIA was scaled back, prices for terrorism risk coverage have declined or remained stable. In some sense, we have conducted a market experiment under TRIA that has illustrated that the private sector is capable of taking on increasing amounts of terrorism risk as the federal government's role recedes. TRIA has generally been effective in encouraging the greater provision of terrorism risk insurance, while at the same time encouraging and supporting private market development. However, by providing a terrorism risk reinsurance without any up-front premiums, it may also have displaced some private sector alternatives. As has been clear from the outset, TRIA was designed as a temporary program. A permanent or long-term federal subsidy of free federal reinsurance was never intended. We firmly believe that temporary programs should be just that – temporary. Given the success achieved under TRIA to date, the obvious question is should the federal government maintain a limited role in the provision of terrorism risk insurance? It is clear that some challenges remain in the market for terrorism risk insurance almost five years after the passage of TRIA and nearly six years after September 11. Insurers have made great strides in modeling loss exposure and managing their concentration of risk; however, the ability of the insurance industry to model the frequency of terrorism attacks is uncertain, and market participants are skeptical of their current reliability. As a result, insurers are cautious in allocating more capacity to terrorism risk, although it appears that gradual increases have been occurring over time. If TRIA were to expire, our general view is that the market for terrorism risk insurance in much of the country would largely be unaffected, but that there could be some dislocations in certain markets and industries. Based on where the market for terrorism risk insurance is today, our view is that TRIA should be phased out in order to increase private sector participation. The following three elements are critical if TRIA is to be reauthorized for a second time: the program remains temporary and short-term; private sector retentions are increased; and there is no expansion of the program. Unfortunately, H.R. 2761 does not meet these critical elements. It is important that the program remain temporary and short-term. When the President signed TRIA in 2002 he said that it should be temporary, and the Administration maintains this position. Given the positive market developments during the last five years under a TRIA program where the federal role has been scaled-back further and further each year, we clearly do not believe the federal government's role in terrorism risk insurance should be made permanent. Similarly, if the program were extended for a long period of time there would be less urgency surrounding the development of private sector solutions, which would lead to market complacency. In considering the length of any extension we must maintain incentives for industry participants to continue to improve their systems (e.g., modeling) and develop private market capacity and innovative solutions. We believe the ten year extension in H.R. 2761 is not consistent with the critical element of keeping the program temporary and short-term. It is also important to continue the trend of increasing the private sector's participation and reducing the role of the Federal Government. Private sector retentions provide financial incentives for insurers to encourage their policyholders to mitigate risk through such measures as improved physical security and evacuation and business continuation planning. Private sector retentions can be increased through deductibles, co-shares, or program triggers. Any extension of TRIA should not backtrack from current levels, but rather should reflect some real amount of increased private sector participation. As has been demonstrated by the increased willingness of insurance companies to take on terrorism risk exposure during the life of TRIA, there is ample opportunity to continue increasing private sector retentions. In addition, recent increases in the capacity of property and casualty insurers, as evidenced by growing surplus and profit levels as well as increased reinsurance availability, should allow for greater private sector retentions. Unfortunately, a number of provisions in H.R. 2761 move away from requiring increased private sector participation, such as leaving insurer deductibles and co-payment amounts flat and unchanged, lowering the program trigger level, and lowering retentions for subsequent events through a reset mechanism. Treasury would oppose these provisions as they are inconsistent with phasing out TRIA and encouraging private provision of terrorism risk insurance – which is the fundamental goal of TRIA. The program should not be expanded to introduce new lines or types of coverage willingly provided by the private market. For example, we do not see any evidence of problems in the market for group life insurance or in coverage for domestic terrorism. These markets continue to function despite not having access to the TRIA program. Expanding the TRIA program to include additional coverage for well functioning markets – as H.R. 2761 proposes – is inconsistent with the appropriate role of the federal government in the terrorism risk insurance market. Treasury would oppose any such efforts that move the program in the wrong direction. Finally, there have been questions raised about the lack of coverage for CNBR terrorism risks. As noted previously, outside of workers' compensation insurance, coverage for CNBR risk has generally not been provided by insurers. However, TRIA does provide coverage for CNBR risk if insurers include such coverage in their policies. If policyholders were to demand CNBR coverage and were willing to pay appropriate prices, we would expect some additional capacity to emerge for CNBR risks. At this time the lack of CNBR coverage does not appear to be leading to any disruptions or imposing any broader costs on our Nation's overall economy. We do not support H.R. 2761's expansion of TRIA's "make available" provision that would require insurers to offer coverage for CNBR risks or its provisions that would lower insurer retentions. Nevertheless, outside the debate surrounding TRIA, we should continue to consider the potential economic implications associated with the limited amount of CNBR terrorism risk insurance coverage that is currently being provided. Conclusion We appreciate the efforts of the Chairman and Members of the Subcommittee in evaluating issues associated with terrorism risk insurance and TRIA. Unfortunately, the risk of terrorism is likely to remain a part of our lives for some time to come, but that is precisely why the federal government needs to encourage the development of the most creative and cost effective means of covering terrorism risks. The most efficient, lowest cost, and most innovative methods of providing terrorism risk insurance will come from the private sector. TRIA should be phased out in order to increase private sector participation. The three critical elements that we have set forth surrounding an acceptable extension of TRIA – (1) the program remain temporary and short-term; (2) private sector retentions are increased; and (3) there is no expansion of the program – reflect the positive experience under TRIA to date, and are grounded in the basic principle of limited government involvement in private markets. Without these critical elements, we would not be supportive of extending TRIA as, in our view, the program would be moving in the wrong direction. H.R. 2761 does not meet our objectives. In Treasury's view, from both a market and economic perspective, it would be better to have no TRIA than a bad TRIA. We are willing to continue to work with Congress toward finding an appropriately balanced solution and to establish the appropriate increases in private sector participation. We look forward to continuing to work with Congress on this important issue. Thank you. I look forward to answering your questions.