News Column

Senate Banking, Housing and Urban Affairs Committee Hearing

February 25, 2014

Chairman Johnson, Ranking Member Crapo, and Members of the Committee, thank you for the opportunity to appear before you today on behalf of The Hartford Financial Services Group (The Hartford) and our property-casualty insurance trade association, the American Insurance Association (AIA), to discuss the important issue of terrorism risk insurance. My name is Doug Elliot and I am President of Commercial Markets for The Hartford. Founded over 200 years ago, The Hartford is one of our nation's oldest insurance companies, among the largest commercial property-casualty insurers, and an insurance partner to over one million small businesses across the United States.

As president of The Hartford's commercial property-casualty lines business, I am responsible for the company's small commercial, middle market and specialty casualty businesses, as well as group benefits. In this capacity, I believe that I can offer an important perspective on the unique challenges of insuring terrorism risk, the market-stabilizing effect of TRIA, and the adverse consequences should Congress fail to maintain the program for the foreseeable future.

Most importantly, as we approach TRIA's expiration at the end of this year, the industry understands and welcomes a healthy reexamination of its merits. TRIA has been a successful and important economic tool. It establishes the right public-private balance of responsibilities and loss-sharing for the United States, promoting national security and an orderly economic recovery in the wake of catastrophic terrorism. While we do not support any changes to TRIA's public-private partnership, we would be happy to work with lawmakers to evaluate the feasibility of changes or modifications designed to improve the overall efficacy of the Act. But in this process, we propose that any potential modifications to the program should be assessed in light of the balance the Act currently achieves. A number of proposals that have been discussed could - in the name of increasing private market capacity for terrorism risk - actually lead the industry to a tipping point beyond which individual insurers would need to make difficult decisions to safeguard a company's financial condition instead of maintaining the current level of exposure to catastrophic terrorism risk. This could result in upsetting the public-private partnership and undermining important sectors of the economy that depend on the availability of terrorism risk insurance, such as construction, real estate, manufacturing, infrastructure and small business generally.

The Insurance Industry's Response to September 11, 2001

It has been more than 12 years since the tragic attack of September 11, 2001. That event forced all Americans to confront directly the previously unforeseen realities associated with a catastrophic terrorist attack on U.S. soil - quite literally, to face a new form of war. Despite the unanticipated nature of the event, The Hartford and other insurers responded to September 11 claims in an unwavering manner and without a single dollar of federal assistance.

However, the devastating economic consequences of the attack forced insurers and other businesses to re-examine the nature of terrorism-related risks, as well as to review how such risks were being spread and managed.

In addition to the incalculable cost of almost 3,000 lives, in today's dollars, claims paid by insurers to their policyholders from September 11 eventually totaled some $32.5 billion dollars--$42.1 billion in 2012 dollars. n1 The Hartford's share of this loss was approximately 3% to 3.5%, as we helped our policyholders recover from the tragic loss. Of course, a large portion of the insured industry loss was effectively reinsured, and the reinsurance industry honored its obligations.

Unfortunately, in the aftermath of the attack, the reinsurance markets withdrew new capacity and the reinsurance market for terrorism evaporated. Without the ability to spread and diversify these risks globally through reinsurance and with no ability to price the risk of terrorism, insurance companies were unable to provide adequate terrorism coverage to commercial policyholders. The effects of this chain of events trickled down to lenders and the construction industry, putting a significant drag on the economy. To support the economy and allow private markets to stabilize, Congress stepped forward in bipartisan collaboration and passed TRIA. TRIA provides a federal backstop to insurance companies for large certified terrorism events above a $100 million loss, while requiring insurers to "make available" (offer) terrorism insurance to commercial policyholders for such coverage as workers' compensation, business interruption and property insurance.

Under the current program, insurers are required to pay insured terrorism losses equal to 20 percent of their entire premium for covered commercial property-casualty lines of business before the government steps in to pay its share of loss. Even then, TRIA requires each insurer to pay 15 cents on every dollar of loss above its deductible. TRIA requires the private sector to absorb at least $27.5 billion in insured terrorism losses before taxpayers are exposed, and then provides a recoupment mechanism to permit the recovery of federal dollars that are expended up to the program's annual $100 billion cap.

By virtue of this post-event, public-private "shared loss" mechanism that preserves significant industry "skin in the game" and only accesses federal dollars for catastrophic terrorism losses, TRIA has effectively established a solvency safety net. This safety net provides the certainty and stability necessary for individual insurers to understand and manage their potential exposure to losses attributable to terrorism attacks, while providing a cap on the potential loss to capital from such an attack. Put another way, far from "crowding out" private market capacity, TRIA's structure creates the environment in which a private terrorism insurance market can exist and function.

In the event of a future terrorist attack, TRIA ensures that private insurance payments flow to those affected businesses that have purchased coverage, as well as to their employees, which in turn helps businesses and the economy recover. These payments will be crucial to minimizing the economic, psychological, and social fallout from an attack. The industry responded effectively to the tragic events in Boston and has the capacity to address single site conventional attacks in the future if they happen. At the same time, if an attack is so massive that it triggers the federal protection established by TRIA, as noted, government payments can ultimately be recaptured through a recoupment mechanism that was established in the legislation. This greatly mitigates any taxpayer costs of this federal program.

It is important to emphasize that taxpayers are protected at every step under TRIA. First, they benefit from the economic security that insurance coverage provides before an attack. Second, after an attack occurs, the immediate flow of claims payments from insurers provides stability and minimizes economic disruption to those who suffer from the attack directly, as well as to all Americans. And finally, in the event of a catastrophic terrorist attack that triggers the government program, any dispersed federal funds can ultimately be repaid through TRIA's recoupment mechanism. Thus, TRIA is both a sensible and indispensable component of national economic security.

The Unique Challenges of Insuring Terrorism Risk

A public-private solution is necessary for the risk of terrorism because it is fundamentally different from other exposures. Private insurance markets are founded on the ability to compile relevant data to (a) measure the likelihood and potential severity of loss to a policyholder for any specific peril and then (b) effectively pool the loss experience across many policyholders exposed to relatively homogeneous, random and independent risks. Quite the opposite, terrorism involves an intentional act carried out at the direction of individual actors and groups with the explicit intention of maximizing overall loss of life and property, as well as economic disruption across as many insureds as feasible. Quite simply, a terrorist attack is not a fortuitous event. Furthermore, terrorism exposure lacks a broad-based spread of risk. Terrorists can pick the target, change the target to bypass security and loss mitigation, and coordinate an attack on multiple targets in diverse locations. The adage - "where there's a will, there's a way" - is particularly appropriate for terrorism risk and effectively neutralizes private mitigation efforts.

Equally important, much of the information regarding terrorist plans and potential targets comes from national security data that is appropriately of limited availability to the public. Insurers therefore lack any sound informational basis for assessing the likelihood or probability of a major terrorist attack. While insurers can price insurance when the nature of the risk is estimable but highly uncertain, ex ante (before the event) insurance mechanisms fail when there is no credible basis for assessing the likelihood of an event.

The potential magnitude or severity of large scale terrorist attacks, particularly those that involve the use of unconventional weapons involving nuclear, biological, chemical, and radiological (NBCR) agents, is largely unknown given the fortunate dearth of prior experience. While insurers have managed loss aggregations for most "conventional" attack modes under TRIA, the industry has limited information on managing exposures to wide-area loss event scenarios that would be the hallmark of NBCR attacks.

The challenges associated with wide-area NBCR terrorism are also manifest in the newest form of unconventional terrorist threat - cyber-terrorism. Under a cyber attack, the origin of the attack can be from any single location where there is a computer and access to the internet. However, the ultimate victims of the attack can be numbered in the thousands or millions, can be widespread geographically, and can be located in any area of the United States. In this setting, traditional insurance company means of exposure management are challenged.

Given the concentration of insured lives and property values in business centers and the unique nature of cyber-terrorism, the risk of wide-area terrorism attacks poses a real solvency threat to insurers -- a threat that can easily eclipse that of natural disasters given the stated intention of a terrorist to maximize economic damage and disruption.

Limited Risk Management Tools are Available

Even with the existence of TRIA, insurers' ability to manage terrorism risk is limited. From a coverage perspective, while TRIA requires a mandatory "offer" as a condition for participation, state laws actually mandate coverage for terrorism for certain lines of insurance. For example, in the 49 states that require workers' compensation insurance, insurers may be obligated to cover on-the-job injuries without exclusion, whatever the cause. Further, a number of states (including those with significant business centers) mandate that insurers cover terrorism- created fire losses, even if a policyholder does not purchase terrorism coverage. As a result, while an insurer may exclude NBCR terrorism coverage in some states, losses caused by the fire following an explosion from one of these perils may be covered.

In addition, as noted above, the industry's lack of credible methods for assessing the likelihood of an attack limits our ability to determine an actuarially fair premium. As noted by the most recent 2010 report on terrorism risk insurance market conditions from the President's Working Group on Financial Markets (PWG Report), "despite the reported improvements in modeling to measure an insurer's aggregate loss exposure, the industry remains uncertain about the reliability of probabilistic models to predict frequency and severity of terrorist attacks." n2

Further, reinsurance capacity for terrorism losses is minimal. Unlike primary commercial lines insurers, reinsurers are not subject to the "make available" clause in TRIA, and their appetite for this risk reflects what type of private market might exist absent TRIA. Citing many of the same issues identified above for primary insurance companies, reinsurance companies offer extremely limited capacity for terrorism risk and generally do not offer coverage for terrorist attacks committed with NBCR weapons. According to the 2010 PWG report, reinsurance capacity available for terrorism risk remains in the $6 billion to $10 billion range, n3 an amount that is well below the estimated industry-wide retention figure under TRIA and well below the mandatory recoupment amount of $27.5 billion in insured terrorism losses.

To provide some perspective, The Hartford's 2014 retention under the Terrorism Risk Insurance Program Reauthorization Act (TRIPRA) is approximately $1.2 billion in company losses. With respect to property, terrorism reinsurance of any material amount within this retention is effectively non-existent. In contrast, for natural catastrophe losses, The Hartford's principal corporate catastrophe treaty provides just under $800 million in reinsurance protection in excess of a $350 million deductible. The Hartford has an additional $135 million in hurricane reinsurance protection financed through non-traditional reinsurance markets (e.g., catastrophe bonds). I wish that the reinsurance markets were willing to provide the same capacity for terrorism within our TRIA retention as is available for natural catastrophes. But, as a company that recently conducted a comprehensive request for proposal process, I can tell you that the reinsurance capacity is simply not available. The 2010 PWG report is interesting in that it indicates that the total amount of reinsurance capacity is up slightly from prior studies. The small increase in reinsurance capacity, undoubtedly available to smaller companies, actually demonstrates the value of the TRIA program to "crowd in" additional reinsurance capacity - that is, it provides reinsurers some assurance that the reinsured companies can manage through a large scale event and remain viable trading partners after a loss.

Given these challenges, how do insurance companies manage the risk of terrorism today? The main tool available to manage the risk of terrorism is to limit exposure concentrations in potential "high target areas." If terrorism exposure concentrations get too high relative to surplus, an insurance company could non-renew entire commercial policies to reduce the terrorism exposure - often creating hardships for the underlying policyholders. These exposure concentrations are especially difficult for certain lines of business like workers' compensation and property insurance, including "fire following" coverage in certain states where exclusions for NBCR attacks are not recognized. Over the past 11 years, with the benefit of TRIA, the insurance industry has successfully managed these concentrations of exposure within the TRIA retentions. Policies shed by one company have generally been absorbed by a competitor.

Without TRIA, however, individual insurers would face large uncapped exposure and would face difficult choices about how to manage down exposures relative to capital, including facing decisions on whether or not to non-renew large portions of their commercial policyholder portfolios, especially given the fact that they cannot exclude the peril of terrorism from workers' compensation coverage and property insurance including the "fire following" coverage in a number of states. In fact, a recent report from Marsh outlined steps that insurers are taking to limit their workers' compensation insurance exposure in light of the continued uncertainty of the future of TRIA. Marsh notes, "Because insurers cannot exclude terrorism related losses and employers are required to buy it, the options available to buyers have been reduced and rate increases have accelerated." n4 For the record, we do not believe that this outcome would be in the best interests of our policyholders or the overall economy.

Proposals to Modify TRIA and Alternative Private Market Solutions

As this Committee knows, the expiration of TRIA at the end of this year presents an opportunity to reexamine the program's merits and the feasibility of modifications designed to improve its efficacy. Almost a dozen years into TRIA, there should be no doubt that the program has brought stability to the private market and has enabled insurers to provide capacity despite the unique characteristics of this risk. TRIA has been shown thus far to be a successful partnership among the federal government, insurers, and policyholders to protect the economy in the event of an attack. Thanks to TRIA and its successors, The Hartford has been able to manage our terrorism exposure within acceptable limits while supporting our policyholders' need for terrorism coverage.

Nonetheless, there continue to be calls to modify the level of federal participation as a means of increasing terrorism insurance capacity. Indeed, some have questioned the need for a continued federal role in backstopping the terrorism insurance market - preferring to prod the markets to develop a purely private solution. The premise of this line of argument is that a market solution, even if second best, is still more desirable than a solution that involves federal government participation. As a general approach on many public policy issues, we support this view. Unfortunately, as we have stressed, terrorism risk is the exception to the market-oriented approach. The existence of the public-private shared loss program enables the market to function.

Eliminating TRIA, or altering it in ways that make the federal role meaningless, will not lead to an expansive private market. Instead, it would very likely lead to a model for terrorism where many businesses retain the risk of terrorism and the federal government loses the robust and stable private terrorism insurance market that TRIA has enabled. While it may be true that the first two extensions of TRIA increased insurer retentions and the industry adapted, any further increases incorrectly aimed at growing private market capacity for this risk may take us over the precipice, and result in a decrease in insurers' ability to offer terrorism coverage.

As we have outlined, the industry's ability to take on more terrorism risk is constrained by our limited ability to manage terrorism risk, the availability reinsurance, regulatory rate and policy form restrictions, and the need to protect company solvency. From The Hartford's perspective - one that is shared by AIA - manipulating TRIA's levers will not increase the supply of reinsurance nor will it allow us to exceed our risk concentration limits and rather may only serve to put more of our capital and, therefore, our solvency at risk. Indeed, at least one rating agency has stipulated that companies could experience ratings pressure if their net exposure to terrorism exceeds 20% of capital and surplus. Total surplus, of course, covers all extraordinary events that may be covered under our policies, from hurricanes, earthquakes, and storms to fires and other accidents that are in excess of reserves. It is not just used to cover terrorism loss.

Note that today, for most large commercial insurance companies, retentions under TRIA already average 8% to 12% of total surplus. Approaching that 20% surplus number identified by the rating agencies may be the tipping point that causes insurers to curtail our aggregate exposure to risks of all types, not just terrorism, which could cause a severe and immediate disruption to the economy, reduce the supply of affordable terrorism coverage, and potentially diminish the nation's prospects for an orderly economic recovery in the wake of a catastrophic terrorism event. If we reach that tipping point caused by well-intentioned, but misguided efforts to further increase private market capacity for this risk, the net result could be that terrorism risk could be redistributed to business owners, borrowers, lenders, employees, and - very likely - the federal government itself through post-disaster relief aid.


Since its enactment in 2002, TRIA has been a success. Terrorism insurance is available and affordable throughout the United States, eliminating economic uncertainty and keeping our economy moving as the long recovery finally gathers momentum. TRIA works because it is an effective partnership between the private sector and the federal government -maximizing private market risk bearing and infrastructure while leveraging the government's pooling capabilities for non-insurable risks that align with our national defense policy. Moreover, TRIA has been administered at minimal cost to taxpayers. TRIA is serving as a key element in maintaining an orderly economic recovery should there be another catastrophic terrorist attack on U.S. soil - prepositioning resources to respond to an attack and thereby thwarting a principal objective of terrorism. At the same time, both AIA and The Hartford recognize that any legislation presents opportunities for improvement, especially against a backdrop of continuous change, and we stand ready to work with the Congress and the Administration to evaluate any potential changes to the legislation and their potential impact on the effective balance achieved by TRIA.

n1 Source: Insurance Information Institute, 2012 dollars excluding Victims Compensation Fund.

n2 Report of the President's Working Group on Financial Markets, "Market Conditions for Terrorism Risk Insurance 2010," at p. 18.

n3 Id. at 19.

n4 Market Risk Management Research Briefing, "Pending TRIPRA Expiration Impacts Workers' Compensation Industry", at p. 1 (Jan. 2014).

Read this original document at:

For more stories on investments and markets, please see HispanicBusiness' Finance Channel

Source: Congressional Documents & Publications

Story Tools