The latest catastrophe bond from U.S. primary insurer Allstate has something very unique about it. The Sanders Re Ltd. (Series 2013-1) issuance, through which Allstate hopes to secure at least $350m of reinsurance cover for U.S. hurricane and earthquake risks, features a trigger design that we’ve never seen before and we understand that this is the first time that a cat bond trigger has been structured in such a way.
The Sanders Re cat bond uses an industry loss trigger, with the data on the industry losses provided by Property Claims Services (PCS) from its Catastrophe Series reports. The PCS industry loss estimate data is fed into the model used for the cat bond and an index level is derived against which it is determined whether the losses have been sufficient to breach the deals attachment point and trigger a loss to investors.
The industry loss data is used to derive a per-occurrence index value, which is then compared to attachment points to derive any loss payment amount. The industry loss trigger is, what’s termed, modified as it uses state specific payout factors which are fed into the calculation of the index value.
Now, none of that is unique at all and in fact is a trigger methodology which has been used in a reasonable number of industry loss based catastrophe bond transactions. What’s unique about the Sanders Re cat bond trigger, is that the PCS loss estimate data is not for the entire industry, rather it is for specific lines of business within the industry.
PCS Catastrophe Series industry loss estimate bulletins have always been provided broken down by a number of different factors, including by state, country and also line of business. The lines of business that estimates are reported broken down for include personal property, auto, and commercial property losses. This segmentation of loss data has always been available within the PCS industry loss estimates, but to date no sponsor has taken advantage of them within a catastrophe bond deal.
Allstate has changed that with Sanders Re. The cat bond event calculation process only uses the personal property and automobile line of business losses, along with state based payout factors, to derive an industry loss index and calculate whether it faces a loss. This has never been done before.
This makes a lot of sense as it means the trigger can be more targeted to the underlying book of business, thus removing an element of basis risk within an industry loss trigger for a cat bond. It allows the cover to be more tightly aimed at the exact losses it is designed to cover which could make this approach very popular with sponsors once the Sanders Re deal has gone through.
It also has some benefits for investors as well, we would imagine, offering a degree of diversification within the cat bond market which doesn’t exist to date. It could also allow for automobile loss only cat bonds to be issued, for which we can think of a few large auto insurers who might appreciate that possibility in future.
Gary Kerney, Assistant Vice President at PCS, told us; “We’re excited to see this innovative use of PCS catastrophe loss estimate data. Historically, most users have opted for the overall estimate, which has served the market well. For sponsors and intermediaries seeking more targeted cover, though, the use of state-and line-specific PCS estimates can be quite effective. We’re interested in seeing how this approach is adopted in the future, as well.”
We’re not sure whether this drilled down level of the PCS loss estimate data is used within the trigger of other reinsurance transactions, such as ILWs or other contracts, it is entirely possible that it is.
If Sanders Re comes to market successfully and the deal structure is well received by investors, which so far we understand it to be, it could lead other sponsors to look to this new line-of-business level industry loss trigger which would be positive for the catastrophe bond market as a whole.