Catastrophe bonds are either complete lunacy or an astounding triumph of statistics and probability theory over common sense, depending on your point of view. Either way, they have proved to be a marvelous way for the insurance industry to tap into the deep pockets of the capital markets.
How do they work? The basic idea is beautifully simple. Today, the likelihood of any potential natural catastrophe, be it fire, flood, storms, tornados, or, of course, hurricanes, can be modeled by a specialist catastrophe modeling house. What is modeled is the probability spread associated with various levels of damage that might ensue if the event under consideration occurs, with respect to a particular piece of valuable real estate—say the Florida coast.
So there are basically two important probability curves here, one concerning the likelihood of the event taking place at all, and the second curve modeling the potential damage if it occurs. Both these things are of great importance both to the sponsor of the bond and to the investors in the bond. It follows from this that the catastrophe modeling house is absolutely key both to the creation of the bond and to the selling process, since experts from the modeling house have to be on hand to explain the “odds” to potential investors.
From here on in, the construction of the bond is basically a horse-trading exercise. The sponsor would ideally like to be able to shift as much risk as possible to the investor community for the lowest fee possible, while the investors, naturally, would like the highest possible fee for the lowest possible risk. The skill in constructing the bond lies in
- getting the catastrophe modeling right and
- setting it up as an attractive proposition for both sides.
Risk, in this instance, is defined as a “trigger” level of damage. For example if a hurricane or, say, a severe winter storm (to give it a European context) hits a specific region then the sponsor (an insurance company) will settle all damages below a certain figure. Above that figure, the investors in the cat bond lose their money, so the insurance company has effectively capped its losses through the bond.
The really nice “clean” thing about catastrophe bonds is that they are fully collateralized investments. The investors have to put up the bond money right at the outset. This is held in a special purpose vehicle whose only function is to pay out if the damage trigger level is reached. If, as the investors hope, there is no catastrophe, or the damage is below the trigger level, then on the maturity of the bond, which is usually around three years, but might run to five, the investors get their principle back. Meanwhile they enjoy a guaranteed premium income on the bond.
The obvious question, for the naïve spectator, is why would anyone invest in a cat bond, given that by definition it is a risk that the insurance company would rather not have on its books? The answers to this are many and various, but they have to do with
- the size of the premium being offered,
- the investor’s judgment that the risk of the catastrophe occurring is low enough to be tolerable, and
- for reasons of portfolio diversity.
Catastrophes are not particularly correlated with other kinds of risk, such as equity risk. Hurricanes don’t care what the markets are doing.
In Part 2 of this article we will look at the growth of this industry since its inception in the late 1990s and at how cat bonds are likely to fare in 2010.
- Catastrophe bonds, part 2: The outlook for 2010 [blog post]
- Catastrophe Bonds: What They Are and How They Function [checklist]
- How to Manage Emerging Market Risks with Third Party Insurance, by Rod Morris
- Viewpoint: Viral Acharya and Julian Franks, Regulation after the Crash
- Viewpoint: Sir John Stuttard, Days of Reckoning
Tags: capital markets , catastrophe bonds , insurance , losses