Climate change, insurability of large-scale disasters, and the emerging liability challenge

University of Pennsylvania Law Review

Volumn 155, Issue 6, 2007, Pages 1795-1842

  Kunreuther, Howard C ^a   Michel Kerjan, Erwann O ^a  

^a UNIVERSITY OF PENNSYLVANIA   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 34548057819     PISSN: 00419907     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Conference Paper

References (118)

1. WORLD ECONOMIC FORUM, GLOBAL RISKS 2007, at 4 (2007), available at http://www.weforum.org/ pdf/CSI/Global_Risks_2007.pdf.
2. NICHOLAS STERN, STERN REVIEW: THE ECONOMICS OF CLIMATE CHANGE i (2007). A prepublication version of the Stern Review is available at http://www.hm-treasury.gov.uk/indepenent_reviews/ stern_review_economics_climate_change/stern_review_report.cfm.
3. Id.
4. Climate Group Interview with John Coomber, former CEO of Swiss Reinsurance Co., available at http://www.swissre.com/internet/pwswpspr. nsf/fmBookMarkFrameSet?ReadForm&BM=../vwAllbyIDKeyLu/rscd-6necs4? OpenDocument (last visited May 1, 2007).
5. AM. INT'L. GROUP, AIG'S POLICY AND PROGRAMS ON ENVIRONMENT AND CLIMATE CHANGE 1 (2006), available at http://media.corporate-ir.net/ media_files/irol/76/76115/aig_climate_change_updated.pdf;
6. see also AIG Adopts Policy on Climate Change, INS. J., May 17, 2006, available at http://www.insurancejournal.com/news/ international/2006/05/17/68491.htm.
7. SWISS REINSURANCE CO., SIGMA, NO. 1, NATURAL CATASTROPHES AND MAN-MADE DISASTERS IN 2004, at 3 (2005), available at http://www.swissre.com/Internet/pwsfilpr.nsf/vwFilebyIDKEYLu/MPDL-6A2D64/ $FILE/sigmal_2005_e.pdf.
8. SWISS REINSURANCE CO., SIGMA, NO. 2, NATURAL CATASTROPHES AND MAN-MADE DISASTERS 2005 (2006), at 3, available at http://www.swissre.com/Internet/pwsfilpr.nsf/vwFilebyIDKEYLu/EWAL-6MBJQ2/ $FILE/Sigma2_2006_e.pdf.
9. Id.
10. There is a growing literature studying the evolution of insured losses due to major catastrophes in the world. Total figures differ from one source to another, mainly due to divergent definitions of a "catastrophe," yet all reach the same conclusion. For example, natural disasters inflicting insured losses above $38.7 million or total losses above $77.5 million are considered a major catastrophe by Swiss Re (we use this threshold in Figure 1). Id. at 38. Munich Re uses a higher threshold. As a result, most figures used in the literature regarding the evolution of catastrophe losses underestimate the real effect on insurers. In the United States, for example, Property Claim Services (PCS) defines a "catastrophe" as an event that inflicts insured losses above $25 million; smaller events (even those that are repeated) are not included in the PCS catastrophe database.
11. See ISO Props., Inc., PCS Catastrophe Serial Numbers, http://www.iso.com/products/2800/prod2802.html (last visited May 1, 2007) ("PCS now defines catastrophes as events that cause $25 million or more in direct insured losses to property and that affect a significant number of policyholders and insurers."). Nonetheless, all studies of which we are aware conclude that insured catastrophe losses have rapidly increased in recent years.
12. This figure is indexed to 2005 dollars. September 11, 2001 insured losses include property and business interruption. See Erwann O. Michel-Kerjan & Howard C. Kunreuther, Improving Homeland Security in the Wake of Large-Scale Disasters: Would Risk-Based All-Hazard Disaster Insurance Help in the Post-Katrina World?, at slide 7 (Aug. 18-19, 2006), available at http://www.usc.edu/dept/create/assets/003/51953.pdf. These data are from the Wharton Risk Center, Swiss Re, and the Insurance Information Institute.
13. The two most costly industrial catastrophes in regards to insurance claims over this thirty-five-year period are the explosion of the offshore oil platform Piper Alfa in the United Kingdom in July 1988, which cost about $3.2 billion to insurers, and the explosion of a petrochemical plant in the United States in October 1989, which inflicted insured damage of $2 billion (in 2004 dollars). SWISS REINSURANCE CO. 2004, supra note 6, at 34 tbl.9.
14. CATASTROPHE MODELING: A NEW APPROACH TO MANAGING RISK 25 (Patricia Grossi & Howard Kunreuther eds., 2005).
15. Disasters that occurred between 2001 and 2005 are shaded in gray. See WHARTON RISK CENTER, MANAGING LARGE-SCALE RISKS IN A NEW ERA OF CATASTROPHES: REPORT ON PHASE I OF THE STUDY 5 tbl.1.1 (2007), available at http://opim.wharton.upenn. edu/risk/library/Report_on_Phase_I.pdf. Data are from the Wharton Risk Center, Swiss Re, and the Insurance Information Institute. Asterisks indicate that the figure excludes $2 to $3 billion in offshore energy losses.
16. Total economic losses due to Hurricane Katrina are still not definitive.
17. MUNICH RE GROUP, TOPICS GEO 2004 14-15 (2005), available at http://www.munichre.com/publications/302- 04321_en.pdf?rdm?457.
18. Dataset obtained from Munich Re. As discussed in note 9 supra, figures differ from one source to another. Munich Re and Swiss Re, the two leading reinsurers in the world, do not use the same definition of catastrophe. Typically, Munich Re's estimates are lower than Swiss Re's, the former selecting only very large natural catastrophes, whereas the latter is less restrictive in its definition. These differences can be very important. For example, when Munich Re estimated insured loss at about $35 billion in 2004, Swiss Re's estimate was $49 billion. More generally, the Swiss Re and Munich Re data underestimate the actual total cost of natural disasters. However, they also constitute the most extensive (both in time and in scale) approximation of that sort.
19. OECD, which stands for Organisation for Economic Co-operation and Development, is an international organization in which the thirty members, democratic countries in the world with the highest income, brainstorm ways to address the issues surrounding globalization. Organisation for Economic Co-operation and Development, http://www.oecd.org (last visited May 1, 2007).
20. SWISS REINSURANCE CO. 2004, supra note 6, at 35.
21. Earthquakes are a major source of catastrophes as well. While not related to climate change, they can also seriously affect insurance capacity. A major earthquake in California today would be economically devastating and have a large impact on how the insurance industry could handle future catastrophes, including weather-related ones. A major earthquake in Tokyo could inflict economic losses at the trillion dollar level, with a critical impact on financial markets worldwide, and even lead to a recession.
22. ASS'N OF BRITISH INSURERS, FINANCIAL RISKS OF CLIMATE CHANGE 16 (2005), available at http://www.abi.org.uk/Display/File/Child/506/ Financial_Risks_of_Climate_Change.pdf.
23. EVAN MILLS & EUGENE LECOMTE, FROM RISK TO OPPORTUNITY: HOW INSURERS CAN PROACTIVELY AND PROFITABLY MANAGE CLIMATE CHANGE 1 (2006), available at http://www.ceres.org/pub/docs/Ceres_Insurance_Climate_Report_082206.pdf.
24. Although not weather-related, the experience of the Kobe earthquake in 1995 highlights the potential for real cataclysms in the Tokyo region. An even bigger earthquake in Greater Tokyo could inflict economic losses in the range of $1 to $3 trillion.
25. U.S. BUREAU OF THE CENSUS, RESIDENT POPULATION AND APPORTIONMENT OF THE U.S. HOUSE OF REPRESENTATIVES: FLORIDA 1 (2000), available at www.census.gov/dmd/www/resapport/states/florida.pdf;
26. OFFICE OF ECON. & DEMOGRAPHIC RESEARCH, THE FLA. LEGISLATURE, FLORIDA DEMOGRAPHIC SUMMARY, available at http://edr.state.fl.us/population/popsummary.pdf.
27. Andrew Dlugolecki, Thoughts About the Impact of Climate Change on Insurance Claims, in WORKSHOP ON CLIMATE CHANGE AND DISASTER LOSSES: UNDERSTANDING AND ATTRIBUTING TRENDS AND PROJECTIONS 54, 66 (Peter Höppe & Roger Pielke, Jr. eds., 2006), available at http://sciencepolicy.colorado.edu/sparc/research/projects/extreme_events/ munich_workshop/full_workshop_report.pdf.
28. The NFIP is a public insurance program created in 1968, in which insurers play the role of intermediaries between the policyholders and the federal government. Following Hurricane Katrina, the program had to borrow $20 billion from the federal government in 2006 to meet its claims. Congress is considering substantially modifying the program.
29. MUNICH RE GROUP, TOPICS 2000: NATURAL CATASTROPHES-THE CURRENT POSITION 91 (1999).
30. Robert P. Hartwig, Ins. Info. Inst., Hurricane Season of 2005: Impacts on U.S. P/C Insurance Markets in 2006 & Beyond, at slide 53 (Mar. 2006) available at http://server.iii.org/yy_obj_data/binary/744082_1_0/katrina. ppt.
31. Erwann Michel-Kerjan, Energy Infrastructure: Coping with Disasters, WHARTON RISK MGMT. REV., Fall 2005, at 7, available at http://opim.wharton.upenn.edu/risk/review/ Fall2005.pdf.
32. It is fair to say that the impact on the energy infrastructure would have been even more devastating had a major hurricane hit a city like Houston, Texas. Such a scenario would result in large-scale property and business interruption losses for oil and chemical industry firms and their insurers.
33. Kerry Emanuel, Increasing Destructiveness of Tropical Cyclones over the Past 30 Years, 436 NATURE 686, 686 (2005).
34. Id. at 687.
35. Id.
36. Category 4 hurricanes have sustained winds from 131 to 155 miles per hour; Category 5 systems, such as Hurricane Katrina at its peak over the Gulf of Mexico, have sustained winds of 156 mph or more. See NATIONAL WEATHER SERVICE, NATIONAL WEATHER SERVICE MANUAL: TROPICAL CYCLONE DEFINITIONS, 3-4 (2006), available at http://www.weather.gov/ directives/sym/pd01006004curr.pdf (describing the Saffir-Simpson Hurricane Scale).
37. P.J. Webster et al., Changes in Tropical Cyclone Number, Duration, and Intensity in a Warming Environment, 309 SCIENCE 1844, 1846 (2005).
38. Id. at 1846 (footnote omitted). This significant increase in observed tropical cyclone intensities has also been linked to warming sea surface temperatures in another recent study.
39. See C.D. Hoyos et al., Deconvolution of the Factors Contributing to the Increase in Global Hurricane Intensity, 312 SCIENCE 94, 96 (2006) ("The implication of these results is that the strong increasing trend in [Category 4 and 5 cyclones] for the period 1970-2004 is directly linked to the trend in tropical [sea surface temperature], and that other aspects of the tropical environment, although they influence shorter-term variations in hurricane intensity, do not contribute significantly to the global trend of increasing hurricane intensity.").
40. Indeed, a spirited debate persists on this issue. Compare Christopher W. Landsea, Hurricanes and Global Warming, 438 NATURE E11, E11-E12 (2005) (questioning the data used in Emanuel, supra note 30, by suggesting that earlier estimates of hurricane wind speed may have underestimated the true figures),
41. with Roger Pielke, Jr. et al., Are There Trends in Hurricane Destruction?, 438 NATURE E11 (2005) (criticizing the index proposed in Emanuel, supra note 30, as a "misleading" indication of destructiveness),
42. and Johnny C.L. Chan, Comment on "Changes in Tropical Cyclone Number, Duration, and Intensity in a Warming Environment", 311 SCIENCE 1713b (2006) (questioning the conclusion of Webster, supra note 34, instead suggesting that factors such as wind shear are responsible for the observed increase in cyclone activity, and determining that "it is difficult to conclude that more intense typhoons are likely to occur in a warmer world"),
43. and with P.J. Webster et al., Response to Comment on "Changes in Tropical Cyclone Number, Duration, and Intensity in a Warming Environment", 311 SCIENCE 1713c (2006) (contending that the factors identified by Chan, supra, such as wind shear and moist static energy, are not correlated with hurricane intensity in a statistically significant manner).
44. Christopher W. Landsea et al., Can We Detect Trends in Extreme Tropical Cyclones?, 313 SCIENCE 452, 452 (2006) (footnotes omitted).
45. Patrick J. Michaels, Is the Sky Really Falling? A Review of Recent Global Warming Scare Stories, 576 POLICY ANALYSIS 1, 17 (2006).
46. Id. at 18-19.
47. For more discussion on this issue, see generally Dlugolecki, supra note 24,
48. and Evan Mills, Insurance in a Climate of Change, 309 SCIENCE 1040 (2005).
49. See supra note 12 and accompanying text.
50. Consider, for example, insurance for property damage caused by hurricanes. An insurer's expected losses are relatively low, because in a typical year, the policy-holder will not suffer a hurricane loss. However, it is possible that losses will be quite high-far in excess of those expected at the time policies are priced. In the event of a serious hurricane, a substantial portion of the loss must be paid from insurer capital. For other extreme events, such as terrorism, maximum losses are extremely high relative to expected losses, so the capital issue is critical.
51. The variance for a single loss L with probability p is Lp(1-p). If L = $100,000 and p = 1/1000, then Lp(1-p) = $100,000(1/1000) (999/1000) or $99.90.
52. See Colin Camerer & Martin Weber, Recent Developments in Modeling Preferences: Uncertainty and Ambiguity, 5 J. RISK & UNCERTAINTY 325, 325-26 (1992) (discussing situations in which individuals are unable to characterize ambiguous risks).
53. We would like to thank Jack Nicholson of the Florida Hurricane Catastrophe Fund (FHCF) for providing us with the FHCF database and Patricia Grossi and Robert Muir-Wood of Risk Management Solutions (RMS) for their analyses of the FHCF data. They provided us with the relevant EP curves and tables presented in this section.
54. For purposes of this Article, we use the new generation of models developed by RMS in 2006.
55. MANAGING LARGE-SCALE RISKS, supra note 13, at 25 tbl.2.2.
56. Id. at 26 fig.2.8.
57. Press Release, American Re Publishes Topics "Annual Review of North American Natural Catastrophes 2001" (May 8, 2002), available at http://www.munichreamerica.com/content/press/topics_05-8-02.htm.
58. See supra notes 6-9 and accompanying text.
59. See James M. Stone, A Theory of Capacity and the Insurance of Catastrophe Risks, pts. 1&2, 40 J. RISK & INS. 231, 339 (1973).
60. Stone also introduces a constraint regarding the stability of the insurer's operation. See Stone (pt. 1), supra note 51, at 232 (recognizing the stability constraint and suggesting it results from investors' and managers' preferences for "some regularity"). However, insurers have traditionally not focused on this constraint in the context of catastrophic risks.
61. See MARTIN F. GRACE ET AL., CATASTROPHE INSURANCE: CONSUMER DEMAND, MARKETS AND REGULATION 62-64 (2003) (noting that Florida imposed a moratorium on policy terminations after Hurricane Andrew and discussing the pools set up by the state to insure those unable to obtain coverage in the voluntary market).
62. Richard J. Roth, Jr., Earthquake Insurance Protection in California, in PAYING THE PRICE: THE STATUS AND ROLE OF INSURANCE AGAINST NATURAL DISASTERS IN THE UNITED STATES 67, 69-78 (Howard Kunreuther & Richard J. Roth, Sr. eds., 1998).
63. Howard Kunreuther et al., Ambiguity and Underwriter Decision Processes, 26 J. ECON. BEHAV. & ORG. 337, 342-44 (1995).
64. Howard Kunreuther, Insurability Conditions and the Supply of Coverage, in PAYING THE PRICE, supra note 54, at 17, 34 tbl.2-1. Ratios are based on mean premiums across number of respondents for each scenario. N = number of respondents.
65. Laure Cabantous, Ambiguity Aversion in the Field of Insurance: Insurers' Attitude to Imprecise and Conflicting Probability Estimates, 62 THEORY & DECISION 219, 235 (2007).
66. For a survey of adverse selection issues, see Georges Dionne et al., Adverse Selection in Insurance Markets, in HANDBOOK OF INSURANCE 185 (Georges Dionne ed., 2000).
67. See Dominique Henriet & Erwann O. Michel-Kerjan, Optimal Risk-Sharing Under Dual Asymmetry of Both Information and Market Power: A Unifying Approach 20 (June 7, 2006) (unpublished working paper, available at http://_opim.wharton.upenn.edu/risk/library/06-10.pdf) ("[W]hen the insurer is better informed than the insured about risk, then whatever his position (principal or agent), that is whatever his bargaining power, low risk is better insured than high risk.").
68. Pat West, Lessons learned from Southern California's Firestorms Released, FIRE CHIEF, Jan. 23, 2004, available at http://firechief.com/news/firefighting_lessons_learned_southern.
69. Theo Francis, CEO Says Allstate Adjusts Storm Plan, WALL ST. J., Sept. 6, 2005, at Cl.
70. MANAGING LARGE-SCALE RISKS, supra note 13, at 105.
71. See Regulators Establish Task Force on Climate Change, INS. J., Mar. 14, 2006, available at http://www. insurancejournal.com/news/national/2006/03/14/66451/htm?print=1 (reporting the establishment of the task force and its objectives).
72. Press Release, Risk Mgmt. Solutions, New RMS View of U.S. Hurricane Activity Rates Increases Losses by 40% in Fla. and Gulf Coast (Mar. 23, 2006), available at http://www.rms.com/NewsPress/PR_2006_0323_HUActivityRates. asp.
73. See supra Part II.
74. A.M. BEST, METHODOLOGY: CATASTROPHE ANALYSIS IN A.M. BEST RATINGS 2-3 (2006), available at http://www.ambest.com/ratings/methodology/catastrophe.pdf.
75. Some states may, however, appeal this judgment.
76. Michael Faure & David Grimeaud, Principles of Liability: A Theoretical Framework, in DETERRENCE, INSURABILITY, AND COMPENSATION IN ENVIRONMENTAL LIABILITY 19-20 (Michael Faure ed., 2003).
77. STEVEN SHAVELL, ECONOMIC ANALYSIS OF ACCIDENT LAW 32-46 (1987).
78. 127 S. Ct. 1438 (2007).
79. 42 U.S.C. § 7521 (a) (1) (2000).
80. Id.
81. Massachusetts v. EPA, 415 F.3d 50, 58 (D.C. Cir. 2005), rev'd, 127 S. Ct. 1438 (2007).
82. See id. at 64-74 (contending that Massachusetts has standing, the EPA has the authority to regulate GHG emissions, and that the EPA's decision not to do so failed to meet the "arbitrary and capricious" standard).
83. Id. at 60. For an additional analysis of Massachusetts v. EPA,
84. see Daniel A. Farber, Uncertainty as a Basis for Standing, 33 HOFSTRA L. REV. 1123, 1128 (2005).
85. 28 U.S.C. § 1350 (2000).
86. See J. Kevin Healy & Jeffrey M. Tapick, Climate Change: It's Not Just a Policy Issue for Corporate Counsel-It's a Legal Problem, 29 COLUM. J. ENVTL. L. 89, 101-02 (2004).
87. See Id. at 102.
88. Elizabeth E. Hancock, Note, Red Dawn, Blue Thunder, Purple Rain: Corporate Risk of Liability for Global Climate Change and the SEC Disclosure Dilemma, 17 GEO. INT'L ENVTL. L. REV. 233, 244 (2005).
89. David A. Grossman, Warming Up to a Not-So-Radical Idea: Tort-Based Climate Change Litigation, 28 COLUM. J. ENVTL. L. 1, 24 (2003).
90. Myles Allen, Liability for Climate Change: Will It Ever Be Possible To Sue Anyone for Damaging the Climate?, 421 NATURE 891, 891 (2003).
91. See id. at 891-92 (describng "apportion[ing] liability according to the change in risk" as the "equitable solution" to compensating victims of climate change);
92. see also Myles R. Allen & Richard Lord, The Blame Game: Who Will Pay for the Damaging Consequences of Climate Change?, 432 NATURE 551, 552 (2004) (observing that English law permits increased risks to serve as a basis for compensation).
93. Stephen W. Lagakos & Frederick Mosteller, Assigned Shares in Compensation for
94. Radiation-Related Cancers, 6 RISK ANALYSIS 345, 346-47 (1986).
95. See supra notes 35-39 and accompanying text.
96. There might be positive externalities associated with global climate change as well. For example, it has been shown that, under certain assumptions, climate change on U.S. agricultural land will increase annual profits by $1.3 billion (in 2002 dollars), or 4%. See Olivier Deschenes & Michael Greenstone, The Economic Impacts of Climate Change: Evidence from Agricultural Output and Random Fluctuations in Weather, 97 AM. ECON. REV. 354, 354 (2007).
97. Joni Hersch & W. Kip Viscusi, Allocating Responsibility for the Failure of Global Warming Policies, 155 U. PA. L. REV. 1657, 1657 (2007).
98. See, e.g., Smith v. Van Gorkom, 488 A.2d 858 (Del. 1985) (holding directors personally liable for breaching their duty of care in their approval of a merger without adequately informing themselves of the terms of the merger).
99. The two largest insurance companies in the American D&O market are Chubb and AIG.
100. Letter from the Investor Network on Climate Risk to Christopher Cox, Chairman, SEC (June 14, 2006), available at http://www.ceres.org/pub/ docs/Ceres_INCR_SEC_letter_061406.pdf.
101. CERES, GLOBAL FRAMEWORK FOR CLIMATE RISK DISCLOSURE (2006), available at http://www.ceres.org/pub/docs/Framework.pdf.
102. MICHELLE CHAN-FISHEL, FIFTH SURVEY OF CLIMATE CHANGE DISCLOSURE IN SEC FILINGS OF AUTOMOBILE, INSURANCE, OIL & GAS, PETROCHEMICAL, AND UTILITIES COMPANIES 23 (2006), available at http://www.foe.org/camps/intl/ SECFinalReportandAppendices.pdf.
103. Id. at 10-13.
104. Kenneth E. Anderson & Donna Ferrara, Disclosing Environmental Liabilities: Director, Officer, and Insurance Issues, 16 ENVTL. CLAIMS J. 3, 11 (2004).
105. E-mails from Swiss Re (Nov. 23 & 24, 2006) (on file with authors).
106. Marsh Inc., Climate Change: Business Risks and Solutions, RISK ALERT 23 (2006), available at http://solutions. marsh.com/climates/.
107. Healy & Tapick, supra note 77, at 102-03.
108. Grossman, supra note 80, at 17-18 (citations omitted).
109. ALLIANZ GROUP & WORLD WILDLIFE FUND, CLIMATE CHANGE AND INSURANCE: AN AGENDA FOR ACTION IN THE UNITED STATES 33 (2006).
110. Id.
111. Id. at 33.
112. MILLS & LECOMTE, supra note 21, at 17 (citation omitted).
113. Id.
114. Id.
115. Id.
116. Id. at 20.
117. For a discussion on the data quality challenge, see, for example, Eberhard Faust, Peter Höppe, Angelika Wirtz & Silvio Schmidt, Trends in Natural Catastrophes - Potential Role of Climate Change, in WORKSHOP ON CLIMATE CHANGE AND DISASTER LOSSES: UNDERSTANDING AND ATTRIBUTING TRENDS AND PROJECTIONS, supra note 24, at 89, 89.
118. Swiss Re, Munich Re, and Leuven University have developed time series on some disaster data, but they differ in their scope and are usually not public information. Workshop Summary Report, in WORKSHOP ON CLIMATE CHANGE AND DISASTER LOSSES: UNDERSTANDING AND ATTRIBUTING TRENDS AND PROJECTIONS, supra note 24, at 4, 5.