Portal de Periódicos da CAPES

Catastrophe Management in a Changing World: The Case of Hurricanes

2008; Wiley; Volume: 11; Issue: 2 Linguagem: Inglês

10.1111/j.1540-6296.2008.00140.x

1540-6296

Richard A. Derrig, Jay S. Fishman, Martin F. Grace, Joan T. Schmit,

Risk Management in Financial Firms

This article features a presentation and discussant comments on hurricane and wind insurance organized by Richard A. Derrig for the American Risk and Insurance Association (ARIA) 2007 Annual Meeting in Quebec City, Quebec, Canada. The moderator, Richard A. Derrig, is President of OPAL Consulting LLC, Providence, RI. Richard formed OPAL after retiring in 2004 from the Auto Insurers and Insurance Fraud Bureaus of Massachusetts. He has published numerous articles on auto insurance, fraud, and other important insurance issues in risk and insurance and actuarial journals. The principal presenter is Jay S. Fishman, Chairman and Chief Executive Officer of The Travelers Companies, Inc., a Fortune 100 company with assets in excess of $100 billion and revenues in excess of $25 billion. Travelers offers a wide range of property and casualty insurance products and services to businesses, organizations, and individuals in the United States and selected international markets. Mr. Fishman was named President and Chief Executive Officer of Travelers in 1998 and Chairman in 2000. From early 2000 until October 2001, he also served as Chief Operating Officer of Finance and Risk for Citigroup, as head of Citigroup's global insurance businesses and head of its consumer business in Japan and Western Europe. Mr. Fishman is an alumnus of the University of Pennsylvania, having graduated with a bachelor's degree in economics, magna cum laude, in 1974 and with a master's degree in accounting from the Wharton School, also in 1974. At the University of Pennsylvania, he is a member of the Board of Trustees, the Board of Overseers of the Graduate School of Education, and the Industry Advisory Board of the Financial Institutions Center for the Wharton School. He serves on the Board of Directors of the National Academy Foundation. Also, he is an active member of the Business Council, a Vice Chairman of the Kennedy Center Corporate Fund Board in Washington, D.C., and a past Chairman of the American Insurance Association. Our two discussants are Professors Joan Schmit and Martin Grace. Joan holds the American Family Insurance Chair in Risk Management and Insurance at the University of Wisconsin-Madison, where she has been on the faculty since 1988. Currently, she is Chair of the Actuarial Science, Risk Management, and Insurance Department in the School of Business. She also holds a visiting appointment with the University of St. Gallen. Dr. Schmit has been active in several important professional associations. She has been an academic moderator for the International Insurance Society and on the Research Committee of that organization. She is a past President of ARIA, the Insurance Society, and the Risk Theory Society. She has served on the board of the Griffith Foundation and is a Research Fellow at the China Center for Insurance and Social Security Research at Peking University. The second discussant, Professor Grace, is currently the Associate Director of Research at the Center for Risk Management and Insurance at Georgia State University. He is an Associate in the Andrew Young School of Public Policy Studies and holds a law degree and a Ph.D. in economics from the University of Florida. Dr. Grace's research focuses on public policy, regulation and taxation, and his research has been published in many economics, insurance, and public policy journals. He is a former President of the Risk Theory Society and a current Associate Editor of the Journal of Risk and Insurance. Richard A. Derrig (Moderator): The focus of today's discussion concerns the tail of loss distributions. The tail is associated with catastrophes, or cat risk, such as the devastation from Hurricanes Katrina, Wilma, Andrew, and others.1 Catastrophes such as these are not just a U.S. issue, but an international issue. Jay Fishman will provide his perspectives on the effect of wind in catastrophic losses while Professors Schmit and Grace will follow with discussions that put wind catastrophes in general insurance contexts. Jay S. Fishman: As we gather here today, the insurance industry is at a crossroads regarding hurricane wind insurance. We, at Travelers, are suggesting that any sustainable solution to this challenge should be provided by the private insurance market, which is ultimately best equipped to handle this exposure. We have a conceptual framework for our proposal that I will share with you today. It is the beginning of what we hope will be a collaborative and comprehensive study, leading to the creation of a private, market-based solution to what is a serious economic challenge. The focus on hurricane exposure has been precipitated not only by climate experts' warnings of an increase in the likelihood of severe hurricanes, but also by a very real increase in demand for coastal property—resulting in the past decade's escalating property values and aggressive development. The 2003 movie, Something's Gotta Give, provided me with inspiration for the title of my remarks today and with an illustration of the lure of coastal property. For those who've seen it, I suspect you may remember the setting; many scenes take place in a home presumably in the Hamptons, right on the ocean, and feature Diane Keaton and Jack Nicholson picnicking on the property's beach. It could easily have been Cape Cod, the Florida Keys, or Hilton Head instead. Well, they say life imitates art. Or maybe it's the other way around. In fact, 55 percent of the U.S. population now lives within 50 miles of the coast, and estimates show continued growth—particularly as baby boomers look to their retirement years and envision the next phase of their lives with sand between their toes. And we're not talking about modest beach cottages. Sprawling homes and skyscraper condominiums are replacing smaller beach bungalows. Planned condominium construction in South Miami Beach alone over the next 2 years, for example, includes more than 2,000 units, each costing between half a million dollars on the low end to $16 million on the high end, and with an aggregate property value of $6 billion. As we develop and build on the coast, we continue to be warned by climate experts who indicate that we have entered into a period of warming ocean temperatures that will likely increase the size and frequency of catastrophic storms. So, take today's very expensive, very congested coastal property, add to it the specter of greater frequency and severity of weather—think of the three hurricanes in 2005—Katrina, Rita, and Wilma—and the equation yields a substantial rise in the risk of significant, perhaps unprecedented, catastrophic losses. As a leader in this industry, we believe we have an obligation to bring thoughtful ideas to the table to suggest how to deal with the effects of our changing climate. With that backdrop, let's take a step back and look at what the past several decades have taught us about the business of insuring wind. As this history is reviewed, four key issues will emerge: insurance availability, insurance market stability, insurer profitability, and policyholder affordability. In the 1980s and well into the 1990s, we had a relatively long period of benign weather and experienced few hurricanes—which many believe to be the result of an Atlantic cooling cycle. Prominent exceptions included Hurricane Hugo in 1989, which produced $6.8 billion of damage (in 2006 dollars), and Hurricane Andrew in 1992, which produced $22 billion of damage (in 2006 dollars) and unprecedented 200 mile per hour winds. It is generally believed that Andrew wiped out all of the premiums collected by the industry on property policies in the state of Florida for all of the years leading up to it. (As a side note, insurers providing homeowners coverage in Louisiana and Mississippi are expected to pay claims in an amount equal to all homeowners' insurance premiums paid in the state since 1981 and 1989, respectively.) As a result of Hurricane Andrew, about a dozen insurers went out of business, which led others to worry about their own company's solvency. Insurers began to reassess their coastal exposures, evaluating the additional amount of capital needed to support the increased risk. This led to price increases in order to maintain a minimum level of profitability. State policymakers became concerned. Influenced by local pressure, some of them established rules and practices mandating that insurance remain available at what now would be considerably suppressed rates, given the significant increase in risk. Insurance companies were faced with a situation where the rules in place before the wind blew—in other words, the rules under which those insurers had based underwriting decisions—were not the same as the rules in place following the storm. Insurers were required to continue offering policies, and thus commit capital, on noneconomically viable terms. This led insurers to be skeptical about writing additional business and committing capital to these areas long term. Price suppression and inconsistent regulations led to the first issue: insurance availability and obtaining the right price for the exposure. Then, between 1992 and 2003, the frequency and severity of named hurricanes in Florida diminished. Questions began to surface about whether 1992's Hurricane Andrew was an aberration. Insurance companies that continued to push for actuarially sound prices, based on a long-term view and with the knowledge that another “Andrew” was possible, had a new problem. Industry critics and some political leaders had pegged us as “fear mongers” or “profiteers.” These critics didn't understand, or perhaps care, how insurance works. Obviously, insurance companies don't mint money. Rather, to pay claims, they must spread the risk of economic loss among as many individuals as possible subject to the same kind of risk and over time. In this case, the industry, recognizing that the risk was greater than previously thought, was trying to price to anticipate another Andrew-type storm and to spread risk over time. So, the second issue was one of perception: the perception that insurance companies were trying to artificially inflate rates and be profiteers at the expense of the public. The latter led to yet a third issue: what's the right price to cover wind for a coastal-exposed property? It had become clear that severe weather losses are predictable only over the long term. While everyone agreed that rates should be fair and equitable, the problem was that rates could appear to be out of whack in any given year and even in any 5 years, and significant losses 1 year could wipe out premiums and profits from several years or more. Let's remember that insurance companies are businesses, and they have to attract private capital to make this all work—and the capital can have a short horizon. So, how could insurance companies set rates and evaluate returns in the short term? The basic affordability of insurance also had become a related question. If actuarially sound rates were to be set, there would be many middle-income and fixed-income residents who would not be able to afford insurance for the coastal properties they owned because their values had increased so tremendously. This brought up very basic concerns about disruption and dislocation. Finally, problem number four is that state government and political intervention, such as the overextension of “residual market pools” has limited competition and increased instability. These pools were initially created as markets of last resort, but local pressures to keep rates low have made them not only the market of first resort but for all practical purposes in some areas, the market of only resort. These and other politically expedient actions artificially constrained the price of coverage and frustrated marketplace attempts to correct the supply–demand imbalance. And the intervention itself can be a problem. In the case of Florida Citizens Property Insurance Corp., for example, we are so far away from economic reality that should a major hurricane hit Florida, Citizens is not sustainable. In an average hurricane season—whatever that means—it is estimated that Florida would experience a total insured loss of $10.2 billion. Under this scenario, the Florida Hurricane CAT Fund would have a surplus of $2.1 billion, and Citizens would have a deficit of $1.9 billion. So, in other words, for an average hurricane season, Citizens would actually lose money. Furthermore, under many scenarios, the company would require a bail-out. This could mean seeking government support or taxation against individuals or businesses. A federal government bail-out would mean that all taxpayers—even those living in Iowa and Wisconsin—would cover the hurricane losses of those living on the coast. Basically, the 45 percent of the population not living on coastal property would subsidize the 55 percent of the population on the coast—what has been referred to in a recent editorial on the subject as “windstorm socialism.” So, any proposed wind insurance solution needs to address these four historical issues—availability, stability, profitability, and affordability—and, at the same time, it needs to respond to the fact that hurricanes don't recognize state borders. We have seen various proposals calling for a federal government role in providing wind insurance—whether as a backstop or as an addition to the federal flood program. We believe that private insurers should continue to play a substantial role in providing coverage for wind. All of this brings us to Travelers' proposed insurance concept for hurricane-prone areas. Our concept is to create a private market Coastal Wind Zone along the Gulf and Atlantic coasts, say from Texas to Maine, in which the federal government would oversee most aspects of wind underwriting, including pricing. In this proposal, there is no federal government financial role. It is not called for, nor is it anticipated. I must note that this proposal is a work in progress. We don't have all the answers, and we are engaging experts to help us think through the related challenges and iron out the details. Overall, we believe that solving this challenge requires collaboration between the insurance industry and local, state and federal governments. This proposal is an attempt to provide a framework to facilitate that collaboration and to sponsor a dialogue to tackle this difficult issue. So, let's review our proposal in light of the four key issues that have plagued private market hurricane wind insurance efforts over the past several decades. With respect to the first two issues regarding availability and stability, Federal oversight along the Atlantic and Gulf coasts would create a more stable environment for insurers and, more importantly, for consumers. Oversight would ensure that rates are actuarially sound and appropriate to cover the volatility of natural catastrophes throughout the defined region. Also, rates would be commensurate with the risks of a particular area within the Coastal Wind Zone so that people living in higher risk areas would pay more than those living in lower risk areas. Insurers would have greater confidence that regulations would be established for the long term, would not fluctuate and change postevent, and would be, therefore, more inclined to underwrite wind exposure in these areas. The federal government would not underwrite the risk, nor would it provide a financial backstop under this program. Rather, the proposed Coastal Wind Zone calls for a federal regulatory oversight role—perhaps a federal regulatory commission. The commission would provide: A single risk-based set of rating rules to determine actuarially sound rates, as well as a single set of underwriting guidelines covering wind exposure in designated coastal areas. Application of a uniform set of laws, regulations, and policy forms that establish the rights and responsibilities of consumers and insurers in the payment of coastal wind claims. States would continue their regulatory oversight over the broad spectrum of insurance activities, including licensing insurers, agents and brokers; enforcing statutory accounting requirements; conducting examinations of insurers' financial positions and market conduct; monitoring insurer solvency in concert with the federal regulatory commission; and administering markets of last resort. Both state policymakers and the National Association of Insurance Commissioners (NAIC) could serve in an advisory capacity to the Coastal Wind Zone Program. The benefits of the uniform program would pass directly to consumers. Consistency of regulation both before and after a catastrophe would decrease uncertainty, allowing competition to increase, which should in turn lead to more efficient pricing. That brings us to the third issue, which was one of profitability and perception. Financial metrics within the insurance marketplace—that is, costs for consumers and profits and losses for insurers—can only be reasonably viewed in the long term. A single catastrophic event or a series of events in 1 year can eliminate profits a company earned over several years. Identifying sound rates, in this circumstance of relatively low frequency and relatively high severity, requires a “cost-plus” dynamic—in other words, the cost of the loss, including loss adjustment expense (LAE), plus a marginal profit. In order to ensure this fairness, we propose the creation of a policy structure within the Coastal Wind Zone that requires insurers to implement prospective premium adjustments to consumers in periods following significant profits or losses. These premium adjustments should help to address the concerns of critics, which we noted earlier. A rolling 10-year time period might be an appropriate basis for this framework, over which time insurers' results would be compared against an appropriate combined ratio range. A portion of the profits would be returned if actual losses incurred during that time period were less than anticipated and assessments would be made if actual losses were more than anticipated. The benefit of this arrangement is a shared management of risk, and the perception of insurance companies “winning” and policyholders “losing” would be eliminated—or at least substantially minimized. Now for the fourth issue—insurance affordability. Let's face it: some coastal residents can afford actuarially sound rates better than others. In Florida, for instance, 18 percent of the state's residents are over age 65, many of them presumably on fixed incomes and with limited assets. To address the issue of affordability, we suggest the creation of temporary federal tax credits for the purchase of insurance to cover wind exposures along the Atlantic and Gulf coasts. The credit would be based upon need, as determined by income and asset level, property value and cost of wind insurance coverage, and would be targeted to those who are least able to afford the insurance. By way of background, under the rules of the current Congress, when a new federal program is suggested, it is incumbent on the group suggesting the program to identify funding—what has been referred to as the “pay-go rules.” This tax credit suggestion is our answer to the question of funding for this program so that no additional funds will be required to implement our proposal. These tax credits would be funded by increases in taxes for those who live in the Coastal Wind Zone and who can afford to pay the taxes. We recognize that this is a transfer of wealth; however, the tax credits would only be available for a fixed period. In other words, while we don't want to cause major dislocation of current coastal residents, everyone would be “on notice” that after a specific number of years—say 10 or 15, tax credits would no longer be available for their insurance. At some point, individuals need to be able to afford their insurance without a subsidy. This proposal goes back to the basic tenet of insurance, which is to spread the risk among as many as possible that are subject to the same kind of risk. The proposal spreads the risks among those who are actually subject to coastal wind exposure, not to those who are not, as does a federal government bailout. A recent survey conducted by the American Consumer Institute found that 84 percent of respondents were unaware of cross-subsidies for homeowner's insurance that benefit coastal properties, and 64 percent disapproved of them. Another subject that has received a great deal of attention in the months following Hurricane Katrina is the topic of “wind versus water.” Discussions centered on what damages were caused by flood, and therefore covered by the National Flood Insurance Program (NFIP), and what damages were caused by wind, and therefore covered by the private wind insurance policy. That has resulted in much debate—and could be the topic of another speech—but for our purposes here, we would suggest that the creation of the Coastal Wind Zone may give us an opportunity to provide more clarity regarding what is covered by the private market through wind insurance and what is covered by the federal flood program. Regardless of how damages are attributed to wind or water, we believe it's important for people who live along the coasts and in flood-prone areas to purchase both flood insurance and wind insurance. In addition, we recognize that responding to a changing climate requires a comprehensive response that will include enforcing stricter building codes, committing to prudent land-use planning, improving construction technology, and acknowledging the true costs of coastal development. As an important aspect of our Coastal Wind Zone insurance proposal, we encourage risk mitigation strategies. Foremost is the adoption and enforcement of building codes that match expected windstorms, for both new construction and renovations. As a resource, the insurance industry created the Institute for Business and Home Safety to assist coastal building code departments in understanding and adopting construction models and materials that withstand natural disasters. The use of storm shutters, impact resistant windows, and other home safety equipment should be encouraged through incentives to lessen potential damage. Under our proposal, coastal states would be entitled to federal grants if they adopt a proposed federal building code and related mitigation programs. Property land use management is another mitigation strategy that would lessen a storm's impact. For instance, coastal wetlands have been shown to minimize a hurricane's impact upon landfall by soaking up moisture and slowing the speed of the storm. When overdeveloped, these areas present an elevated exposure to wind and flood losses during a storm. Zoning regulations that take into account the beneficial nature of wetlands and the proper development of coastal land provide long-term benefit to consumers. With almost $7 trillion of insured coastal property from Texas to Maine, solving the insurance challenge posed by the threat of hurricanes and implementing correct risk control measures are critical issues to the economic future of the United States. We are advocating further study and actively facilitating the necessary dialogue with insurance and governmental leaders to further refine our Coastal Wind Zone idea. We believe this proposal provides a promising concept for offering wind insurance on our coasts that benefits consumers. After all, the primary role of the federal government in regard to coastal wind exposure is to help people prepare for and respond to disasters. The appropriate role of insurers is to help individuals manage their risks and cover their exposures related to their coastal property. By staying true to those roles, we will best be able to assess the true costs of coastal living and facilitate responsible development there. Professor Martin Grace: I am unsure about whether “something's gotta give” or if we can get out of this corner that we have been painted into. What I will talk about is the cat risk environment and several proposals (including the Travelers' proposal) concerning how to solve cat risk problems and then provide a short conclusion. Private insurance markets work well and are competitive in homeowners insurance, even if they are not as efficient as they might be. Hurricane Hugo shocked this market, and the shock of Hurricane Andrew quickly followed. Naturally, a reevaluation of hurricane risk and homeowners' insurance pricing occurred. But reaction to insurer's changing prices was not favorable by outsiders to the industry. So, for example, Florida regulators intervened in the market to solve the affordability and availability problems caused by insurers' reevaluation of homeowners' insurance risk in that state. Had there been no intervention in the Florida homeowners' insurance market, prices would now be higher there: insurer capital would be available for Florida, and a very competitive homeowners' insurance market would exist. In other words, if risk-based pricing had been allowed in Florida in 1993–1995, the problems that now exist in the Florida homeowners' insurance market just wouldn't be present today. Instead, the state of Florida became the biggest insurer in this market in the mid-1990s. Concerned with the level of risk being borne, the state of Florida then began allocating high risk policies to insurers without the financial wherewithal to bear that risk. Inadequately capitalized insurers sprang up which were willing to gamble with bankruptcy in taking on homeowners' risk. The end result was government-induced market failure in Florida's homeowners' insurance market. Today, once again, the state of Florida is the largest insurer in Florida, and its governor is accused of being a socialist for socializing the market for homeowners' insurance. Further, distrust has arisen between insurers and regulators and between customers and regulators. For example, insurance regulators and the governor promised price reductions that were not forthcoming. Part of the problem is that customers can't distinguish between risk-based price (based on spreading of risk principles) and profiteering. So this is how we became painted into a corner with respect to this market. Although I believe that if a free market had been able to operate 15 years ago we would not have a problem now that opportunity is not on the table at this time. Thus the Travelers' proposal is probably a second best solution. But the Travelers' proposal is not the only proposal circulating. Let's review some of these other proposals. One proposal is for the federal government to accept all cat risks, and insurers would then just write homeowners' insurance to cover losses except those relating to cat risk. This would solve the problem of insurers charging a premium for homeowners' insurance that is unacceptable to all of its constituents, including regulators and customers. Another proposal would have the federal government underwrite wind and flood risk jointly and operate similarly to the National Flood Program. The advantage of putting both coverages together is that it avoids disputes about whether a loss was caused by wind or water, and customers would clearly understand what their actual coverage is. Some consumer advocates' proposals call for some or all of the following: insurers are required to cover all cat risks without any sort of bailout from state or federal governments, insurers are required not to charge high prices or earn high profits, insurers are required not to use hurricane projection models to forecast losses, or insurers are required not to exit the market. Yet another proposal exists which I term, “Consumer Advocates ‘Hypothesized’ Proposal B.” Under this proposal, the state completely replaces the private homeowners' insurance market by underwriting this risk itself. This proposal is appealing to all those who don't trust the insurance industry and believe that the state can do a better job. Certainly the latter sentiment has been expressed many times in Florida. The Travelers' proposal has several advantages over all of the proposals mentioned above. It sets similar rules across all states and calls for similar methods for resolving disputes through the federally regulated wind pool. It removes the incentives for states to vacillate between requiring low prices in some periods and expanded coverage in other periods. Another positive, innovative aspect of the Travelers' proposal is the potential consumer/insurer “profit sharing” if losses are less than expected. While there is no explicit risk to the Federal Treasury, there are tax credits for insurance for those who need it, and these are funded through higher property taxes. Historical experience with respect to tax credits is that they are hard to phase out once they are put in place. Thus, it is reasonable to believe that this proposal would result in a net transfer of wealth that will result in higher property taxes. Thus, this is a disadvantage. On balance, however, I believe that the Travelers' proposal is a reasonable one given the circumstances the homeowners' insurance market is in. It represents an attempt to make cat risks more transparent, thereby building trust with consumers. It eliminates incentives for states to expropriate ex ante through price regulation or ex post through contract reinterpretation. One may question whether it will dampen or eliminate price competition for catastrophe coverage. And, of course, there is always the risk that in considering the Travelers' proposal, Congress will make “improvements” to the proposal that make the resulting program legislated very different from the initial proposal. Joan Schmit: A primary purpose for this opening plenary session at our annual meeting is to connect the intellectual efforts of industry and academics. The issues are great, as