The market for catastrophe bonds, bonds designed to protect insurers from the risk of paying out on natural disasters, is poised to grow this year by as much as 30 per cent, according to European insurer Axa. However, to fulfil the growth prospects, cat bonds, most of which are domiciled in Cayman, still have to overcome certain obstacles delegates heard at the Insurance Linked Summit in February.
While Axa believes more issuers are coming to the market for cat bonds looking for an alternative to reinsurance, Andre Perez CEO of insurance manager and consultancy Horseshoe Group, argued at the ILS Summit that cat bonds are still closely held by a too small number of investors for the market to grow significantly.
“That is the holy grail. In my opinion this market is not going to grow unless we expand the investor space.”
It is easy to name the top five investors that are involved in all of the deals, Perez noted. To expand the market to a broader universe of investors, bankers had a role to play, because investor education is still needed in the industry.
“There are large hedge funds and money managers who really know nothing about insurance,” he said.
In addition there are other impediments to market growth, such as the high transaction costs for corporate cat bonds, for example issued by utility providers.
For the issuers the objective of cat bonds is to transfer the risk to the capital markets, in order to access capital that has a lower cost of capital than reinsurance, the alternative form of risk transfer.
One problem, according to Perez, is that all of the investors in cat bonds are also active investors in collateralised reinsurance and thus ideally placed to compare the rates of return in both markets.
Investors, like pension funds, buy cat bonds because of their low correlation with other financial instruments, the economy and other risks. However, investors risk losing all of their investment if a defined catastrophe occurs.
While most investors consider cat bonds a buy and hold investment, Insurance Linked Security Funds value the liquidity of cat bonds, which are listed and tradable in a secondary market, over illiquid collateralised reinsurance. However, to improve the liquidity of the secondary market, the market needs more cat bonds and a broader investor base.
Steve Britton, who oversees Horseshoe’s Cayman office, observed that of the three main markets – cat bonds, industry loss warranty index trades and collateralised reinsurance – in 2010 the collateralised reinsurance market overtook the cat bond market in size.
According to Swiss Re Capital Markets the total amount of catastrophe bonds outstanding in 2011 stood at $13.6 billion. The private collateralised reinsurance market is estimated to be in the region of $15 to $20 billion, whereas ILWs make up around $5 to $6 billion, Britton said.
Although investors’ prefer more information, some are thinking about collateralised deals because they are tailored risk, said Britton.
The state of the market
2011 was the year with the highest catastrophe related economic losses of $350 billion and it would have been the most costly year for the insurance industry, if Japan had been more fully insured. Thus the $108 billion of insurance losses were the second highest amount after 2005 with Hurricane Katrina when insured losses were $123 billion.
In 2011 the market saw 27 cat bond deals totalling $4.3 billion. This was just short of the expectation but still the fourth largest year for ILS issuance, Britton said.
Historically, over 90 per cent of all cat bond deals have been domiciled in Cayman. But of the 27 deals in 2011, six were domiciled in Ireland, eight in Bermuda and 13 in the Cayman Islands, signifying that other jurisdictions are catching up. In addition Cayman attracted only one new issuer, with the remainder coming from repeat issuers. Bermuda in contrast was able to attract all new issuers, Britton said, who added Bermuda had a natural advantage in that large insurers already go there to buy reinsurance, which makes it a logical step to think about cat bonds in the jurisdiction as well, Britton said.
Attracting more reinsurers to Cayman would therefore be a good idea to round out the financial centre, he concluded.
Indemnity versus non-indemnity
The payout structure of the deals shows that the majority of last year’s cat bonds featured indemnity triggers, which are related to the issuer’s actual losses in the same way as if they had purchased reinsurance. In contrast, non-indemnity triggers are either related to the hazard, such as wind speeds in the case of a hurricane, to overall industry losses or to specific models.
Investors prefer non-indemnity because it is transparent and it is quick resolution, said Britton. Deal sponsors, in turn, prefer indemnity because there is no basis risk on the insurance they are buying.
For the market to grow, Britton said, it needs to see more indemnity type of transactions.
The art of pricing
Another area that makes the cat bonds market unique is pricing.
The pricing of bonds and insurance valuation is still a little bit of a mysterious thing, said Perez. “You really don’t know how people eventually decide how to price that bond.”
“In the ILS universe you have the very sophisticated buyers that are going to look at the modelling that are going to put that into their own aggregation platform and look at the marginal capital allocation and see how that returns. And those are among the most sophisticated ILS investors,” Perez said.
“And then you have some of them and they say well you know what is my expected loss again? OK what is the spread over my expected loss? OK sounds good. Let’s buy. So I don’t think there is a real science to it.”
When RMS, one of the top two providers of hurricane risk models, altered its modelling criteria in 2011, resulting in a higher estimate of hurricane risk, the extent of the modelled risk increase caught the market by surprise.
Perez said that while RMS may have been right to make the adjustments to its model, “the timing was bad and the magnitude was terrible”.
As a result RMS’ competitor AIR appears to have gained some market share, as it was the preferred model used for the majority of the recent cat bonds. “Most of [the issuers are] modelling AIR because they get lower numbers, lower estimates and lower costs on the deal,” agreed Britton.
“Investors though are smarter than that. They use the RMS model and what we are seeing on the traditional side is that that market has priced more towards the RMS model, whereas the cat bond market may have not yet incorporated it as much.”
It is important to remember that hurricane risk models are only a tool, said Perez. Sophisticated investors would always run their own internal models to the extent that they have the data, but for the less sophisticated investors the impact of the model change was huge, because they rely on those models, he said.
“People understand it is an art as much as science. It is not exact.”