Risk Securitizations and Insurance-Linked Notes: Where Capital Markets Meet Insurance
Book: Structured Finance and Insurance: The ART of Managing Capital and Risk Author: Christopher L. Culp Publisher: Wiley Finance, 2006 ISBN: 978-0-471-70631-1
Chapter 22 marks the transition from Part Three (Structured Finance) into Part Four (Structured Insurance and Alternative Risk Transfer). And the bridge product is perfect: insurance-linked notes, or ILNs. These are securities whose principal and interest payments depend on the occurrence of insured events. They sit right at the intersection of capital markets and insurance.
In a typical asset securitization, you issue bonds backed by loans or receivables. In a risk securitization, you issue bonds to fund an insurance contract. The goal isn’t raising money for the sponsor. The goal is managing risk. The SPE sells insurance (or reinsurance) to the sponsor, and investors in the notes are the ones bearing the risk. If the insured event happens, note holders lose principal and interest. If it doesn’t, they collect a nice premium.
The Basic Structure
Most ILNs look like a fully funded synthetic CDO. An SPE is established, bankruptcy remote from and unconsolidated with the sponsor. The SPE is licensed as an insurer or reinsurer. It issues notes, gives the proceeds to a trustee, and the trustee invests in low-risk securities. The sponsor pays insurance premium to the SPE, which the trustee adds to the collateral pool.
If a covered event occurs, the trustee liquidates collateral to pay claims, and note holders lose some or all of their principal. If no event occurs, note holders get their principal back plus interest (LIBOR plus a spread funded by the insurance premium and investment income). An asset swap converts investment income into LIBOR-based cash flows to match the floating-rate liabilities.
Cat Bonds: Synthetic Reinsurance
The most famous ILNs are cat bonds, sponsored by insurance or reinsurance companies to create additional reinsurance or retrocession capacity. For years, cat bonds were purely a capacity play. Traditional reinsurance was cheaper, so insurers only went to the capital market when they couldn’t get enough coverage any other way. But by the time of this book, cat bond spreads were falling and they were starting to compete on price too.
Triggers and Payouts
How a cat bond defines and measures “the bad thing happened” is crucial. There are four main approaches.
Full indemnity: Claims are based on actual losses paid by the insurer. Most transparent for the insurer, but investors worry about moral hazard.
Indexed: The trigger and payment are based on an external loss index (like PCS regional catastrophe indexes). Eliminates moral hazard concerns but introduces basis risk. The index might not track the sponsor’s actual losses very well.
Parametric: The trigger is defined by objective physical parameters. For earthquakes: epicenter location and magnitude. For hurricanes: landfall point and storm category. Very fast to pay out (no loss development period needed) and easy to hedge. But basis risk can be significant.
Modeled loss: Parameters are fed into a model that estimates expected losses for the specific insurer. A middle ground between indemnity (too much moral hazard) and indexed (too much basis risk). Increasingly popular.
Examples from the Reinsurance World
SR Earthquake Fund (Swiss Re, 1997): $137 million in four classes of notes linked to California earthquake losses via PCS index. First two classes were the first cat bonds rated investment grade. Class C was a pure binary bet: 100 percent principal loss if PCS-reported California quake losses exceeded $12 billion. From a financial engineering perspective, each class is a coupon bond plus embedded digital options.
Parametric Re (Tokio Marine, 1997): Earthquake reinsurance for Tokyo quakes with a parametric trigger based on JMA scale and location. An inner grid and outer grid around Tokyo determined payouts. A 7.4 JMA quake in the outer grid meant a 44 percent principal holdback. Same magnitude in the inner grid: 70 percent.
Residential Re (USAA, 1997): One of the best-known cat bonds. Eastern U.S. hurricanes. $477 million raised, nearly four times the planned amount. True indemnity coverage: 80 percent of $500 million XS $1 billion of actual USAA losses. Class A bonds (AAA) had principal protected. Class B (BB) had principal at risk.
Georgetown Re (St. Paul, 1996): Used both notes and preferred stock. Notes got principal protection through zero-coupon Treasuries (AAA rated). Preferred stock had no protection. The reinsurance was a classical 10-year proportional treaty on catastrophic XOL business. By including both underwriting and investment results, St. Paul made the ceded business well diversified.
Winterthur Hail Bonds (1997): No SPE at all. Winterthur issued subordinated convertible bonds directly with “WinCat coupons” linked to motor vehicle hail and storm claims. If the number of claims hit 6,000 or more, the coupons reset to zero. Otherwise, the coupon was one-third higher than normal convertibles. Risk transfer was estimated at only CHF 9 million, but it was considered a success.
Beyond Cat Risk
Cat bonds expanded into other insurance lines as spreads fell. The SECTRS deal (1999) created synthetic retrocession for Gerling Credit Insurance Group’s European trade credit reinsurance. Three classes of securities against a portfolio of 92,000 businesses.
Life insurance got involved too. American Skandia securitized future mortality and expense charges on life policies. National Provident Institution securitized the future surplus on a block of life policies. Hannover Re used a Dublin SPE to finance European life reinsurance expansion.
Swiss Re’s Vita Capital (2003) was the first securitization of excess mortality risk. $250 million of notes with principal at risk if a European mortality index exceeded 130 percent of its base level. At 150 percent, note holders lost everything.
And in 2005, Oil Casualty Insurance Ltd. did the first casualty risk securitization through Avalon Re: $405 million in three tranches.
Corporate Risk Securitizations
The cat bond concept migrated to corporate risk management in the mid-1990s. Corporations started buying insurance directly from the capital market.
Oriental Land/Tokyo Disneyland (1999): The first corporate cat bond. Parametric trigger for earthquakes near the theme park. The deal had two parts: a $100 million risk transfer piece through Concentric Ltd. (insurance that pays based on quake magnitude and proximity) and a $100 million contingent debt piece through Circle Maihama (post-loss financing activated by the same trigger). A quake of 7.5 on the JMA scale within 10 km of the park meant 100 percent payout and total principal loss for investors.
Vivendi Universal/Studio Re (2002): $175 million to insure Universal Studios California against earthquake damage, using modeled loss triggers covering property damage, workers’ comp, and business interruption.
Toyota Motor Credit/Grammercy Place (residual value insurance): This one is different. TMCC insured the risk that returned lease vehicles would be worth less than their residual value. $566 million in three classes of notes. Each year’s coverage included a 10 percent co-pay and a deductible. Class A notes at risk only if car values fell more than 23 percent below expectations. Class C notes at risk if values fell 9 percent.
Freddie Mac/MODERNs: Mortgage default insurance through a securitization. Five classes of notes whose principal was recalculated each period to reflect only non-defaulted mortgages in a $15 billion reference pool.
EDF/Pylon (2004): Transmission and distribution asset insurance for France’s national power company after devastating 1999 windstorms. Two tranches, the first covering a first storm and the second triggered only by a second storm within five years. This was notable because it insured a core business risk. Insurance companies get nervous about that because of moral hazard and adverse selection. The parametric trigger helped manage those concerns.
FIFA/Golden Goal (2003): $260 million in event cancellation insurance for the 2006 World Cup. The trigger was specific: if no winner could be determined before August 31, 2007. The trigger excluded world war and player strikes but included terrorism. Notes were issued in three currencies and naturally hedged through currency-matched assets and liabilities.
Cat Swaps and Exchange-Traded Derivatives
The chapter closes with derivatives alternatives. Cat swaps are floating-for-floating swaps where one reinsurer makes insurance-linked payments to another following a triggering event. Swiss Re and Mitsui Marine did one in 1998 with a parametric Tokyo earthquake trigger.
Risk swaps let reinsurers literally swap catastrophic exposures. Swiss Re and Tokio Marine exchanged California earthquake risk for Japanese earthquake risk, Florida hurricane risk for Japanese typhoon risk, and French storm risk for Japanese cyclone risk. Three tranches at $150 million each under a single master agreement.
Exchange-traded cat derivatives (CBOT futures and options) were tried multiple times but never gained traction. Cat futures launched in 1992 and effectively died within a few years. PCS options replaced them in 1995 but suffered the same fate. The trading volumes were embarrassingly small. On one side, you had a concentrated reinsurance industry that didn’t need standardized contracts. On the other, the off-exchange market hadn’t developed enough to create demand for standardized products. Culp suggests these products might succeed someday, just not yet.
My Take
This chapter is packed with examples, and each one illustrates a slightly different application of the same basic technology. What I find most striking is how versatile the ILN structure is. Earthquakes, hurricanes, hail damage, trade credit defaults, mortality rates, car residual values, mortgage defaults, windstorm damage to power lines, World Cup cancellation. Same SPE, same trustee, same asset swap. Different insurance policy underneath.
The corporate ILNs are particularly interesting because they represent companies going directly to the capital market for insurance, bypassing the traditional insurance industry entirely. When EDF couldn’t find adequate coverage from insurers after those 1999 windstorms, it went to investors instead. And it worked.
The failure of exchange-traded cat derivatives is a useful counterpoint. Not everything that makes theoretical sense finds a market. Sometimes the existing bilateral market is good enough. Sometimes the standardized product arrives before the market is ready for it.