At War With the Weather: Managing Large-Scale Risks in a New Era of Catastrophes.
Several major natural disasters have occurred in the United States during the last decade. Florida was affected by four severe hurricanes in 2004. In the following year, hurricanes Katrina, Rita, and Wilma caused record losses by taking 1,500 lives and triggering more than $180 billion of damage reimbursements. What are the implications of these destructive forces of nature for the supply and demand for property insurance, what programs should be put in place to address the observed increase in natural disaster risks, and what roles should insurers and other key stakeholders play in these programs? Howard Kunreuther and Erwann Michel-Kerjan and their research team provide answers to these and other important questions. At War With the Weather provides a comprehensive study of how mitigation of risk, insurance, and other risk transfer instruments can be jointly applied to manage the risks of major natural disasters in the United States. This book provides a very valuable contribution to the literature on natural disaster risk management and insurance because of this comprehensive set up. Natural disaster risks in the United States are partly covered by private insurers who, for instance, provide wind damage insurance, and by the federal National Flood Insurance Program (NFIP). This book provides detailed analyses of both by focusing on the states of Florida, New York, Texas, and South Carolina. In my opinion, the authors rightly point out that the framework and analysis applied in this book are of relevance for other countries and can be extended to other low-probability, high-impact risks. In general, the book is written in a way that is accessible for practitioners in the field of insurance and natural disaster risk management, while the material is also of interest for academics. Several chapters of this book could be selected for undergraduate or even graduate courses on risk management and insurance.
The paperback edition of this book starts with an epilogue that reflects upon lessons for risk management that can be drawn from recent disasters, such as the financial crisis, the BP oil spill, the catastrophic earthquakes in Haiti and Chile, and the tsunami in Japan. The authors believe that these disasters are signals of more severe catastrophes in the future. They pose the question: why are we not more proactive in preventing these extreme events? A tentative answer is that empirical evidence, which has been illustrated with several practical examples, indicates that investments in measures that reduce disaster risks are insufficient because we tend to focus attention on immediate problems and use simple heuristics or choice rules to avoid thinking about catastrophic risks. It has been proposed that solutions to this should be found in strategies that promote long-term planning by providing short-term economic incentives. The authors introduce two main proposals for such strategies on which the remainder of the book elaborates. First, the development of multiyear compensation schemes and multiyear (insurance) contracts can provide incentives for long-term planning to the key stakeholders involved in disaster risk management. Second, governments should be more proactive in encouraging firms and individuals to undertake protective measures that reduce catastrophe risk.
A main strength of the book is that it provides both theoretical frameworks of the demand and supply of natural disaster insurance (Chapters 6, 7, and 9) and complementary empirical analyses of unique data sets that have been provided by insurers and the NFIP (Chapters 4 and 10), which have been embedded in the current regulations of insurance markets in the United States. It is these empirical analyses that I found a main asset of the book because they bring alive many of the theoretical issues that the authors put forward. For example, an analysis of 7.5 million NFIP insurance policies in force in Florida in Chapter 4 provides several interesting insights into insurance demand. Many residents in flood-prone areas let their NFIP policy lapse when the contract expires and when they did not experience flood damage. Experience of the NFIP shows that many individuals do not purchase flood insurance even if prices are close to or below the expected value of the loss. Moreover, the book includes results of catastrophe models that estimate the exposure of insurers to natural disaster risks. As an example of how insurers can assess their catastrophe exposure, the authors present an exceedance probability curve for hurricane risk in Florida. The results indicate that there is a 15 percent annual probability that the insured hurricane loss is at least $10 billion and a 5 percent probability that it exceeds $25 billion, which illustrates the huge consequences that hurricanes can have in that state.
The book starts in Chapter 1 with a useful overview of the roles of the key stakeholders involved in mitigating and insuring natural disaster losses. The important influence of insurance regulators on the functioning of insurance markets is extensively discussed in Chapter 2. An interesting example of extensive intervention is Florida where regulators have resisted premium increases that were deemed necessary by insurers who wished to charge premiums that reflect storm risk after the active hurricane season in 2004. As a response, several insurers retreated from the market and aimed to reduce their exposure in some high-risk areas, which resulted in a reduced availability of insurance, as Chapter 3 discusses. Reduced availability of insurance or (perceived) high cost of insurance has put pressure on several states to establish state-run or state-sponsored residual market mechanisms, which in the case of Florida witnessed an unsustainable growth in exposure. Chapter 10 of the book provides interesting empirical results about demand and supply of property insurance in Florida. The results show that insurance provided by Citizens Property Insurance Corporation, the state-run residual market property insurer in Florida, acts as a substitute for private insurance, since the cross-price elasticity indicates that a decrease in the price of Citizens reduces demand for private insurance, albeit to a
small extent. The price elasticity of supply is found to be relatively high: a 10 percent increase in the price that homeowners are willing to pay for coverage results in a 27 percent increase in the number of policies supplied. This implies that a suppression of premiums by regulators can result in a large decrease in supply. A detailed analysis of prices charged in Florida shows that there is a large cross-subsidization of policies, in the sense that premiums in high-risk areas are strongly subsidized by premiums in low-risk areas, which implies that the sensitivity of premiums to risk is flattened out. The price elasticity of demand in Florida is less elastic in high-risk areas, which suggests that premiums could be increased there without the adverse consequence of much lower demand. These findings provide a useful basis for improving the functioning of the property insurance market in Florida and expand the supply of private insurance by allowing insurers to charge premiums that reflect risks.
Chapter 13 provides a series of empirical analyses of the impacts of hurricanes on key stakeholders of disaster insurance schemes. These analyses estimate the ability of insurance to provide coverage against destructive hurricanes and the influence of implementing damage mitigation measures on hurricane risk. The authors estimate the total benefits of implementing mitigation measures to all insured homes, which provides an estimate of the maximum risk reduction that can be achieved through mitigation. This analysis shows that mitigation can reduce losses of a 1/100 year hurricane by 61 percent in Florida, 44 percent in South Carolina, 39 percent in New York, and 34 percent in Texas. If these mitigation measures were to be implemented, and if insurers were allowed to charge risk-based premiums, then the private insurance sector would be able to cover most (if not all) losses from a severe hurricane. The spatial distribution of risk-based premiums shows that charging premiums that reflect risk implies that the premiums of policyholders in coastal communities would increase substantially. These analyses of the benefits of mitigation can provide a good basis for further research on the cost effectiveness of mitigation measures. Unfortunately the book does not provide estimates of the costs of implementing the measures proposed by the authors, while such costs would be an important factor of consideration for policymakers and insurers, as well as for homeowners who often end up paying for mitigating buildings. An important distinction to make is whether mitigation activities are targeted to new or existing buildings. Implementing flood-proofing measures, such as elevation, to existing buildings may be very expensive, while mitigating newly built structures may be a more cost-effective strategy.
The last two chapters of the book close with recommendations for dealing with the problems that have been outlined in detail in the previous chapters. These recommendations are guided by two principles: Principle 1--premiums should reflect risk; Principle 2--equity and affordability should be dealt with. Premiums set below actual hurricane risk in coastal areas have tended to reduce the supply of insurance coverage in such areas, resulted in too much development in high-risk regions, and impaired the ability of insurers to encourage mitigation by providing premium discounts. Stronger incentives for mitigation are important because Chapter 12 of the book shows that individuals rarely invest in mitigation measures even if they are cost effective. The authors, moreover, advocate the introduction of a program of insurance vouchers that is similar to the food stamp program in the United States to assist those households that currently reside in high-risk regions to obtain insurance coverage. Such a program may alleviate problems with the affordability of risk-based premiums, which was examined in Chapter 11. Another main innovation that the authors put forward is to introduce long-term insurance contracts (LTI) with a duration of 5 or 10 years that are tied to a property, instead of the usual annual contracts. The main advantage of LTI is stronger incentives for mitigation, because policyholders are guaranteed that they will receive premium reductions for a long time period (the duration of the contract). Moreover, LTI could be combined with a mitigation loan provided by a bank that helps the policyholder to spread the costs of investments in risk-reducing measures over time. LTI contracts could be applied to property insurance that covers hurricane risk, and could also be applied by the NFIP, as has been proposed in recent articles by the authors Michel-Kerjan (2010) and Michel-Kerjan and Kunreuther (2011).
The potential impacts of climate change on extreme weather and the feasibility of the proposed solutions by the authors for managing disaster risks in the face of the projected increase in natural disaster risk in a changing climate are briefly summarized in the first chapter of the book. Otherwise, the topic of climate change receives minimal attention in the rest of this book. This is an important issue given the challenges that climate change may pose for managing and insuring the risks of natural disasters. For example, the pricing of LTI contracts may constitute a problem because socioeconomic developments and possibly climate change have uncertain effects on future natural disaster risk, which complicates setting premiums that are adequate for the entire duration of the LTI contract. This issue of pricing LTI for flood risk under a range of scenarios of socioeconomic and climate change has recently been examined in a study for the Netherlands (Aerts and Botzen, 2011). Estimates of future risk-based premiums for a variety of contract durations in areas with distinct levels of flood risks show that premiums differ significantly, depending on the expectations of the insurer about the influence of climate or socio-economic change on flood risk. For example, the premiums of a 5-year contract can be considered too low by a large amount if the insurer has a too low expectation of sea-level rise, and vice versa. Given these uncertainties with estimating future risk, insurers are likely to charge a markup on LTI premiums. An important question for further research is whether consumers are willing to pay such a markup.
Aerts, J. C. J. H., and W. J. W. Botzen, 2011, Climate Change Impacts on Pricing Long-Term Flood Insurance: A Comprehensive Study for the Netherlands, Global Environmental Change, 21: 1045-1060.
Michel-Kerjan, E. O., 2010, Catastrophe Economics: The National Flood Insurance Program, Journal of Economic Perspectives, 24(4): 165-168.
Michel-Kerjan, E. O., and H. C. Kunreuther, 2011, Redesigning Flood Insurance, Science, 333: 408-409.
Reviewer: W. J. Wouter Botzen, Assistant Professor, Department of Environmental Economics, Institute for Environmental Studies, VU University Amsterdam, The Netherlands; email@example.com
|Printer friendly Cite/link Email Feedback|
|Author:||Botzen, W.J. Wouter|
|Publication:||Journal of Risk and Insurance|
|Date:||Mar 1, 2012|
|Previous Article:||Index insurance, probabilistic climate forecasts, and production.|
|Next Article:||Form over matter: differences in the incentives to convert using full versus partial demutualization in the U.S. life insurance industry.|