Natural Catastrophes in the Bond Market - A Trader s View Risk Trading Unit Trading risk into value Innsbruck, July 2007 Marcel Grandi 1
Agenda 1. Market development and functional areas 2. Examining the needs of insurers and investors 3. Operating structures assessing required elements 4. Case study 5. Risk trading at Munich Re Agenda 2
1. Market development and functional areas 3
Market development and functional areas in USD bn Munich Re Group Issues Outstanding Cat Bonds Issued Cat Bonds 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 350 300 250 200 150 100 50 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2,7 2,0 2,3 1,8 1,3 1,3 1,1 1,2 1,1 1,0 3,3 1997* 1998 1999 2000 2001 2002 2003 2004 2005 2006 May 2007 1,3 5,1 2,4 5,3 1,3 8,4 2,4 15,0 5,0 2,5 * and prior Source: Goldman Sachs Market Development and functional areas 4
Market development and functional areas Munich Re Track Record Transaction Name Closing Vol.* Covered Perils Carillon Series 2 05/07 150 Hurricane U.S. Nationwide Lakeside Re 12/06 190 Earthquake in California Carillon Series 1 06/06 85 Hurricane U.S. Nationwide Aiolos 11/05 110 Windstorm in selected European Countries Gold Eagle 2001** 04/01 120 Hurricane in East and Golf coast area Earthquake in New Madrid area PRIME CalQuake & EuroWind 12/00 135 Earthquake in N and S California Windstorm in selected European Countries PRIME Hurricane 12/00 165 Hurricane in New York & Miami area Gold Eagle** 12/99 182 Hurricane in East and Golf coast area Earthquake in New Madrid area and California Pacific Re 06/98 80 Typhoon Japan * In USD mn; **Munich Re of America, formerly American Re 5
Market development and functional areas The cat bond market has further matured in 2005, 2006, 2007 Record issuance in 2006 Large pipeline in 2007 New risks transferred to capital markets (industrial 3rd party liability, motor portfolio, trade credit, Mexico EQ, Mediterranian EQ, cat mortality) allow investors to diversify into new risk classes Cat bond spreads widened substantially after Hurricane Katrina after tightening before Katrina Storm resulted in first total loss, i.e. Kamp Re Spread widening in particular for critical exposure zones (e.g. US Hurricane) No detaching of capital markets from reinsurance cycle Strong differences in prices for risks with identical expected loss level Spreads tightening again Market development and functional areas 6
Market development and functional areas Diversified investor market (cat funds, hedge funds, traditional asset managers) Higher risk/return layers placed Competition through Sidecars (private equity) Munich Re reentered the cat bond market European windstorm bond (Aiolos) over EUR 110mn (USD 128mn) Hurricane cat bond shelf program in June 2006 (Carillon) Second series in Mai 2007 (USD 150mn, B rating) Cat bond transaction for Zurich in December 2006 (Lakeside) Market development and functional areas 7
Market development and functional areas Concentration on Peak Exposures Tight capacities High premium level Diversification still limited Traditional reinsurance without competition for diversifying risks Price Capacity Market development and functional areas Pacific NW EQ Australia Wind Australia EQ US Hurricane US EQ Japan Wind Mexico EQ CA EQ Midwest US EQ Europe Wind France Wind Japan EQ 8
Market development and functional areas Investors comfort with regard to owning insurance risk will increase due to Proven track record of past transactions Better model availability leading to reduced asymmetric information Sponsors net benefit will rise due to Lower transaction costs per annum because of increased volumes, multiyear transactions and prevailing best practice methods; use of unissued shelf registrations which allow several takedowns Less risk discounts demanded by investors Decrease in basis risk as modeling capability is improved Even without much interference from outside, the market for capital markets risk transfer solutions is likely to grow further with at least slowly increasing growth rates Market development and functional areas 9
2. Examining the needs of insurers and investors 10
Needs of insurers and investors insurer s view Advantages Capacity in tighter markets Additional and alternative capacity Diversification Collateralization Immediate liquidity Transparency Multi year cover Disadvantages Basis risk Decreases with improvements of modelling capabilities Transaction costs Risk transfer costs Tightening of prices between reinsurance and capital markets for Peak Exposures Complexity Dependence on investor preferences Needs of insurers and investors insurer s view 11
Needs of insurers and investors investor s view Advantages High yield Compared to credit markets Track Record One default Uncorrelated asset class Investment in pure insurance risks Not possible in insurance stocks Rating Traded Cat Bond Spreads BBB+ BB+ BB BB- B+ B Expected Loss 0,23 1,6 2,7 2,8 3,6 8,5 0 1000 2000 3000 Spread max min Needs of insurers and investors insurer s view 12
Needs of insurers and investors investor s view Comparison of BB rated CMBS with a BB rated cat bond (Lakeside) Vergleich Spread CMBS vs. ILS Produkt 650 600 550 500 Spread in BP 450 400 350 300 250 200 150 100 50 0 31.12.2003 31.12.2004 30.12.2005 29.12.2006 Credit Spread CMBS BB Risk Current ILS Bond Spread Needs of insurers and investors insurer s view 13
Needs of insurers and investors issues for investors Liquidity High bid/offer spreads Secondary market write downs (after issuance) Risk assessment Rating and pricing based on risk modelling Different modelling approaches of the Modelling Agencies Modelling arbitrage Change in modelling approach E.g. US Hurricane Doubling of industry losses Downgrade of outstanding Hurricane Bonds Needs of insurers and investors insurer s view 14
Needs of insurers and investors conflict of interest Example US Hurricane MR cat bond 2006 (Carillon I) Investor market Excess demand for hurricane protection first half 2006 Secondary market pricing Write down of new issues Change in investor appetite MR requirements High volume Fair price Multi year cover Whole market protection No exclusions Preference for one year deals Regional covers instead of whole market Exclusions (flood) Needs of insurers and investors insurer s view 15
3. Operating structures assessing required elements 16
Operating structures assessing required elements Efficient protection Costs Compliance with regulations Success of placement Operating structures assessing required elements 17
Operating structures assessing required elements Structural considerations Determine risk to be securitized (peak exposure) Determine expected loss (EL) probability of securitisation Select the EL range offering the best economics (protection vs. price) The lower the EL, the higher the multiple to be paid as spread The higher the EL the lower the multiple Increasing investor appetite for higher EL ranges Fit in overall reinsurance program Operating structures assessing required elements 18
Operating structures assessing required elements Selection of trigger type Parametric (based on event generation) Industry loss (based on damage caused) Modelled loss (reference portfolio) Indemnity (based on financial loss of sponsor) Parametric Modelled Loss Indemnity Modelling Risk Investor Basis Risk Sponsor Operating structures assessing required elements 19
Operating structures assessing required elements Selection of trigger type Indemnity trigger Best for sponsor (no basis risk) Availability of exposure data Publication of exposure data Challenging to sell to investors (moral hazard) Synthetic trigger (parametric, modelled loss, market loss) Wide market acceptance (clarity & transparency) Assessment of basis risk essential Availability of internal know how Operating structures assessing required elements 20
Operating structures assessing required elements Clarity on accounting and taxation Balance sheet consolidation to be avoided Close coordination with auditors No financial leverage Use reinsurer as fronter and legal sponsor Taxation regime No unfavourable taxation of collateral trust assets and premium cash flows (excise tax) SPV location decisive Double tax treaties in place Operating structures assessing required elements 21
Operating structures assessing required elements Selection of service providers Modelling Agency (inc. loss verification) Credibility of model crucial for rating and success of transaction SPV administrator Indenture trustee Legal Limited number of law firms active Placement (one placement bank sufficient) Beauty contest Rating agency (one agency sufficient) Operating structures assessing required elements 22
Operating structures assessing required elements Form of issuance 144A private placement standard Principal at Risk Tranching Currency SPV Off Shore (Cayman or Bermuda, Ireland) Minimal capitalization provided by Charitable Trust Consolidation with sponsor unlikely Operating structures assessing required elements 23
Operating structures assessing required elements Project management (responsible for execution and success) Conceptual design Internal coordination Internal submission Engagement of service providers (organization of beauty contests) Manage internal and external interfaces and service providers Manage costs Control time schedule Controlling Operating structures assessing required elements 24
4. Case Study 25
Case study I Lakeside Re Executive summary Covered Territory: California Risk: Earthquake Modeling Agent: RMS USD 190mn principal at risk bonds 3-year EQ cover for Zurich American Dual Trigger based on PCS industry loss and Zurich ultimate net loss (UNL) Rating S&P: BB+ Spread: 650bps above LIBOR Case study 26
Case study I Lakeside Re Issuer Reinsurer Ceding Insurer Ceding Insurer Affiliates Securities Offered Lakeside Re Ltd., a Cayman Islands exempted company licensed as a Class B insurer Münchener Rückversicherungs-Gesellschaft Aktiengesellschaft ( Munich Re ) Zurich American Insurance Company ( Zurich ), for itself and its Pool Members ( ZAIC Pool ) Certain branches and affiliates of the Zurich Insurance Company other than the ZAIC Pool US$ 190,000,000 Principal At-Risk Variable Rate Notes Closing Date December 20, 2006 Risk Period December 21, 2006 to December 23, 2009 Scheduled Redemption Date December 31, 2009 Triggers (i) Industry Loss threshold based on PCS Reports (ii) Ultimate Net Loss based on actual claims Covered Territory Named Peril S&P Rating Distribution The United States state of California Earthquake On the Closing Date, the Notes are expected to be rated [BB+] 144A Private Placement to Qualified Institutional Buyers in Permitted Jurisdictions who, for U.S. Persons, are also Qualified Purchasers and residents of and purchasing in a Permitted U.S. Jurisdiction or a Permitted Non-U.S. Jurisdiction Case study 27
Case study I Lakeside Re Zurich America $ 190mn Reinsurance Agreement Premiums Munich Re $ 190mn Retrocession Agreement Premiums Deposit Bank $ 190,mn Bank Deposit LIBOR - Eligible Bank Fee Lakeside Re Ltd. Collateral Account $ 190mn Outstanding Principal Amount at Redemption LIBOR + 650bps $ 190mn Note Proceeds $ 190mn Principal At-Risk Variable Rate Noteholders Case study 28
Case study I Lakeside Re For the indemnity layer, the expected loss is 0,43% Industry Trigger Amount Estimated Insured Industry Losses ($ in millions) Annual Probability of Exceedance Estimated Zurich Portfolio Losses ($ in millions) 91.679 0,10% 84.111 0,13% 73.672 0,20% 64.840 0,29% 61.154 0,33% 58.525 0,37% Exhaustion Amount 56.908 0,40% 51.889 0,50% 51.708 0,50% Trigger Amount 45.722 0,67% 40.000 0,88% 37.628 1,00% 32.305 1,33% 25.329 2,00% 14.594 4,00% 3.972 10,00% Case study 29
Case study I Lakeside Re Volume Issue size of US$ 190mn Deal over subscribed Spread 650bps above 3m-LIBOR In line with reinsurance pricing Investors 21 Investors 11% 12% 49% Cat Fund Reinsurer Institutional Hedge Fund 28% Use of Balance Sheet Expiring Redwood V and VI Total bridge financing of US$ 42.5mn Case study 30
Case Study II Carillon 2 Executive summary Covered Area: US East Coast Risk: Hurricanes Modeling Agent: AIR Covered Area Non-Covered Area Case study USD 150mn cat bond for Munich Re Hurricane cover over four wind seasons Trigger based on PCS industry loss (Attachment USD 35 bn; Exhaustion USD 45 bn) Rating S&P: B Spread: 1525bps above LIBOR for 3 2/3 years; seasonalized: 1400bps above LIBOR Attracted new cat bond investors 31
Case Study II Carillon 2 Transaction structure Munich Re Payment Equal to Sum of Interest Spread and Swap Spread Carillon Ltd. Return of Original Principal Amount at Maturity (if no trigger event) Capital Markets Investors AA-/Aa3/AA- (1) PCS Index-triggered US Hurricane Financial Contract Cayman Island SPV Interest: 3M LIBOR + Interest Spread Proceeds (USD) Class E Principal At-Risk Variable Rate Notes Proceeds 3M LIBOR Swap Spread + Principal at Maturity Contingent Security Interest Collateral Account The Bank of New York Trustee 3M LIBOR Swap Spread and Return of par on Permitted Investments Investment Income on Permitted Investments + any gain Notes 1. S&P (Financial Strength), Moody s (Insurance Financial Strength) and Fitch (Insurer Financial Strength) ratings respectively 2. S&P (Long Term Issuer Credit) and Moody s (Senior Unsecured Debt) ratings respectively Morgan Stanley Capital Services Inc. Guaranteed by Morgan Stanley A+/Aa3 (2) Case study 32
Case Study II Carillon 2 Trigger mechanism Principal written down according to Property Claims Services (PCS) loss estimates Generalized Class Payout Functions Principal Reduction (%) 100 Principal Reduction Formula Principal reduction affecting the Notes will be applied linearly between the applicable Event Attachment Amount and the applicable Event Exhaustion Amount using the formula below: 80 60 40 Event Attachment Amount Event Exhaustion Amount Original Principal Amount X Event Amount Event Exhaustion Amount (USD 45bn) Event Attachment Amount (USD 35bn) Event Attachment Amount (USD 35bn) 20 0 PCS Index Any principal reduction is limited to the Outstanding Principal Amount for the Class Case study 33
Case Study II Carillon 2 Risk metrics Class E Probability of Attachment Expected Loss Probability of Exhaustion Base Case 4.61% 3.97% 3.39% Sensitivity Case 5.26% 4.59% 4.00% Notes 1. Source: AIR Worldwide Case study 34
Case Study II Carillon 2 Transaction outcome Volume Issue size of USD 150mn Largest single B rated cat bond ever placed Spread 1525 bps above 3m-LIBOR (annualized) 1400 bps seasonalized Investors 17 Investors 7% 3% 12% 34% 20% 24% Hedge Fund Cat Fund I Bank Reinsurer Pension Fund Insurer 7% 2% 10% 11% 11% 13% 25% 21% Switzerland Cayman Bermuda Germany U.S. U.K. Canada Others Use of Balance Sheet Munich Re bought USD 1.5mn to round up to USD 150mn Case study 35
5. Risk trading at Munich Re 36
Risk Trading Key considerations for Munich Re Attractive market conditions in bn 15 Outstanding Cat Bond Issued Cat Bond 3 Average expected loss* (left axis in %) Pricing multiples * (right axis) 8 7 10 2 6 5 4 5 1 3 2 1 0 0 0 1997* 1998 1999 2000 2001 2002 2003 2004 2005 2006** 2001 2002 2003 2004 2005 2006 * and prior; ** to date *For outstanding cat bonds only Source: Goldman Sachs Source: Lane Financial L.L.C. 2006 Optimal moment to increase our activity in this market segment benefiting from a well developed infrastructure and maturing markets Opportunities for portfolio optimization, increase of capital efficiency and additional earnings with a minimum of launching costs for MR Risk Trading Key considerations for Munich Re 37
Risk Trading Active use of capital markets Munich Re s Risk Trading approach Managing our own risks Optimise portfolio Use of additional capacity Improve our risk/return profile and save costs Managing our clients risks Consulting, structuring, project management and placement support Risk fronting / transformation and (interim) capacity provider Fee and risk-based income Risk warehousing Retain risks Be active player in primary and secondary market Restructuring and reselling Extension of buy and hold strategy Combine and restructure risks Sell at favourable terms and conditions Risk-based, investment and arbitrage income Fee and arbitrage income Risk Trading Active use of capital markets 38
Outlook Dynamic market development Doubling of market volume until 2008 Securitization market 10% of reinsurance market Retrocession predominantly via capital markets New investors (high net worth individuals) New risk classes Risk Linked Securities integral part of risk management spectrum Outlook 39
Contacts Marcel Grandi Senior Manager - Structuring Phone: +49 (0)89 3891-4114 E-mail: mgrandi@munichre.com Münchener Rückversicherungs-Gesellschaft Königinstrasse 107 D-80802 München Conatcts 40
Thank you for your attention! Marcel Grandi 41