Natural Perils and Insurance Quiz Question #1 Which floor in a high rise building should be avoided in an earthquake prone area? 1) First Floor 2) Third Floor 3) Top Floor 4) High rise buildings should be avoided at any cost Page 2 1
Kobe 1995 Page 3 Natural perils losses worldwide on the rise Annual insured losses from 1970 to 2006, in USD bn indexed to 2006 Page 4 Source: Swiss Re sigma 2
Quiz Question #2 How large is the insured loss of Hurricane Andrew in 1992? 1) USD 15bn 2) USD 23bn 3) USD 35bn Page 5 Increasing values in exposed areas Ocean Drive, FL, 1926. Ocean Drive, FL, 2000. Population Growth Rates Page 6 Period 1960-2000 1980-2000 1990-2000 All US 57% 24% 13% Florida 223% 64% 24% 3
The largest loss potentials 110 Peak risks: Earthquake or storm In industrialised countries With high insurance density Page 7 Quake California 50 18 Northridge 1994 45 Hurricane US+Carib. FHCF Katrina 2005 Insurance loss potentials in USD billions: 8 Daria 1990 35 Storm Europe Nat cat events (indexed to 2006) Loss potentials from events with a return period of 200 years (100 years for Hurricane North Atlantic) FHCF: Florida Hurricane Catastrophe Fund state-run JER: Japan Earthquake Reinsurance Scheme schemes Quake Japan Typhoon 20 Japan 8 Mireille 1991 75 JER Tapping into capital markets: Non-Life cat bonds outstanding USD bn 12'000 10'000 8'000 6'000 3'657 8'348 Page 8 4'000 2'000 0 759 1'412 1'812 153 180 714 742 825 1'125 967 990 2'206 1'988 3'102 1'143 3'541 2'138 4'920 2'830 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 As of May 31, 2007 Source: Swiss Re Capital Markets Issued Outstanding 4
Quiz Question #3 How large is the total economic loss for a 200 year earthquake hitting Tokyo? 1) USD 150bn 2) USD 500bn 3) USD 1 000bn 4) Unbelievably high Page 9 Economic vs insured loss potentials (examples) Country Scenario Loss potential (bn USD)* economic insured China EQ >40 <3 Flood >40 <3 Colombia EQ 20 5 Greece EQ 25 4 Italy EQ 35 6 Japan EQ Tokyo 1000 75 Mexico EQ Mex. City 50 6 Netherlands Flood 200 <10 USA EQ California 250 50 USA/Caribbean Hurricane 250 110 * 1 in 200 year events; Hurrican USA/Caribbean 100 years only Source: Swiss Re Nat Cat R&D Page 10 5
Hurricane Katrina Page 11 Hurricane Katrina Page 12 6
Boat Casinos They float Page 13 Boat Casinos they sink Page 14 7
Boat Casinos they fall apart Page 15 Climate change Observations Page 16 Annual Northern hemisphere land surface temperature anomaly (deg C) for the period 1850-2005, relative to the average 1961-10. The error range represents +/-2 standard errors. The smooth curves were created using a 21-point binomial filter to highlight the interdecadal variations. Data from Brohan et al. 2006. 8
Climate change Predictions Page 17 Source: IPCC 2007, Summary for Policy Makers The next 100 years Increased variability Model results Page 18 Source: Nature 2004 9
Hurricane forecasting skills Page 19 Year Forecast Observed Forecast Observed # of Storms # of Storms # Hurricanes # Hurricanes 2001 11 10 6 5 2002 11 12 6 4 2003 13 16 7 7 2004 13 15 8 9 2005 14 27 8 14 2006 15 10 9 5 2007 16? 9? A very recent paper (G. J. Holland, submitted) has demonstrated that current seasonal forecast techniques exhibit only little skill at any time period. Quiz Question #4 Which factor has the biggest impact on the insured natural peril loss of a building? 1) Building quality 2) Surroundings of the building 3) Inhabitants 4) Repair companies 5) Small print of the policy Page 20 10
Impact of building quality Page 21 Residential district in 2050 Page 22 11
Nat Cat modelling: Four elements that determine a cat loss Hazard Vulnerability Value distribution Cover conditions L How often / how strong? How well built and protected? What exactly is covered... where... and how? F Example Hurricane Charley Aug 2004 Sums insured Cover limits Deductibles Exclusions etc. Page 23 Nat Cat modelling: The sophisticated way - simulation Probabilistic Events These artificial events never occurred in history, but they are physically possible and there is no reason why they should not happen one day. Page 24 Note: The displacements etc have been exaggerated for visualization purposes. Gust in m/s 12
Storm surge model Sal. Oppenheim Lounge 28 November 2006 Serge Tröber, Cat Perils Page 25 Simulation of event and damage foot print Page 26 13
Watch where you park your car! Given an event intensity: How are the insured objects affected? Page 27 Cautionary note on forward-looking statements Certain statements and illustrations contained herein are forward-looking. These statements and illustrations provide current expectations of future events based on certain assumptions and include any statement that does not directly relate to a historical fact or current fact. Forward-looking statements typically are identified by words or phrases such as "anticipate", "assume", "believe", "continue", "estimate", "expect", "foresee", "intend", "may increase" and "may fluctuate" and similar expressions or by future or conditional verbs such as "will", "should", "would" and "could". These forward-looking statements involve known and unknown risks, uncertainties and other factors, which may cause Swiss Re's actual results, performance, achievements or prospects to be materially different from any future results, performance, achievements or prospects expressed or implied by such statements. Such factors include, among others: the impact of significant investments, acquisitions or dispositions, and any delays, unexpected costs or other issues experienced in connection with any such transactions, including, in the case of acquisitions, issues arising in connection with integrating acquired operations; cyclicality of the reinsurance industry; changes in general economic conditions, particularly in our core markets; uncertainties in estimating reserves; the performance of financial markets; expected changes in our investment results as a result of the changed composition of our invested assets or changes in our investment policy; the frequency, severity and development of insured claim events; acts of terrorism and acts of war; mortality and morbidity experience; policy renewal and lapse rates; changes in rating agency policies or practices; the lowering or withdrawal of one or more of the financial strength or credit ratings of one or more of our subsidiaries; changes in levels of interest rates; political risks in the countries in which we operate or in which we insure risks; extraordinary events affecting our clients, such as bankruptcies and liquidations; risks associated with implementing our business strategies; changes in currency exchange rates; changes in laws and regulations, including changes in accounting standards and taxation requirements; and changes in competitive pressures. These factors are not exhaustive. We operate in a continually changing environment and new risks emerge continually. Readers are cautioned not to place undue reliance on forward-looking statements. We undertake no obligation to publicly revise or update any forward-looking statements, whether as a result of new information, future events or otherwise. Page 28 14