Newsletter Sign-up

By MATTHEW BRODSKY, senior editor/Web editor of Risk & Insurance®

More than half--10 out of 19--of the most recent catastrophe bond issuances are triggered in one way or another by earthquakes, as revealed by a quick review of the deal directory at Artemis, a Web site dedicated to alternative risk transfer. It got us thinking--that and the recent terrifying destruction in Haiti and Chile ... what would happen to all these bonds, and their investors, should the Big One happen in California, as we all know it will sooner or later.

As for the former question, the answer is not simple.

"Catastrophe bonds are a diverse bunch," said Pascal Karsenti, senior risk consultant at risk modeler AIR Worldwide, which works with issuers to analyze bonds' underlying risks.

Obviously, as mentioned, some bonds are not even exposed to earthquake at all. Those that are might be triggered by a New Madrid temblor, or an international event, not California. Or some could be designed to be triggered by a Northern California quake, others Southern. From one bond to the next, parametric triggers, with thresholds based on measurable phenomena, can be narrowed to focus on different individual fault lines, or set to differing levels of ground-shaking.

The point: Don't assume all CAT bonds will behave as a class just because the underlying general risk is the same.

Of course, if you have a big enough event, said Karsenti, multiple bonds could be triggered.

"That would mean that the CAT bond market has done its job," he said.

Now that leads us to the second question ... when the CAT bond market does its job and pays out to issuers who expected to be covered for a one-in-100-year or one-in-250-year earthquake, what will those investors do when all their principal is vaporized and turned into funds for insurance claims-paying?

Karsenti is confident that investors are comfortable with their exposures. They are licensing the same catastrophe modeling software that (re)insurers use to estimate their exposures when issuing bonds (the same models that Karsenti's employer builds and licenses). Investors are now using the models to help make sure that their bond investments are not correlated (i.e., that they're not all invested in bonds that get triggered by a California quake, for instance).

Of course, some might argue that investors are not quite as advanced at using these models compared with the (re)insurers. Or that much more than modeling goes into understanding the underlying catastrophe exposure, and that if investors don't have the resources or the abilities to carry out this additional research, they could be in for a nasty surprise come quake time.

Glen Daraskevich, senior vice president at catastrophe consulting firm Karen Clark & Co., is generally upbeat about investors' comfort level. "Older" investors in this space know their exposures. And new investors are picking up catastrophe modeling faster than insurance people tend to, he said. That?s because similar sorts of financial modeling software has been used in the investing world for some time. Daraskevich added, though, that investors could have questions, such as why two different model brands put out two completely different loss estimates for the same event on the same set of properties.

"We do spend a lot of time speaking with investors, making sure their expectations are in line," said William Dubinsky, director of insurance-linked securities at Swiss Re. Arrangers of deals need to make sure there is full disclosure and transparency, as well as that they control expectations going in, he said.

The key thing perhaps is, if an earthquake were to happen, investors will want to know what happened and was it contemplated. Say, a 1000-year event occurs underneath Los Angeles.

"That's something investors understand will happen and they will lose their money," Dubinsky said.

If a fault line were to erupt that was not modeled, however, then you could have a "huge problem," he added. Or let's say you have an issuer who expected to collect if a certain event happened and then they don't when the event does occur; you could see issuers get peeved.

But back to the investors ? one additional thing to consider is the scale of the exposures that they're facing when they get into taking on catastrophe risks. As John Seo, managing principal and co-founder of CAT bond investment fund Fermat Capital, explained at a recent industry meeting, the insurance market does scare him with what is considered "a bump in the road."

A $1 billion loss, for instance, is not a bump in the road for an investor.

But then again, it's no secret that CAT bonds are below investment grade, which means they know they can lose their principal. That's why investors earn oversized rewards in the good times, said Chi Hum, managing director at GC Securities.

He reiterated the point made earlier. As long as the Big One is not unexpected, the CAT bond market will function as it ought to and will continue to.

March 1, 2010

Copyright 2010© LRP Publications