Compass Re II cat bond’s parametric trigger explained by Fitch

Rating agency Fitch has provided some more detail and background information on U.S. primary insurer AIG’s latest catastrophe bond, the $200m Compass Re II Ltd. (Series 2015-1) deal which launched earlier this week.

With this Compass Re II 2015-1 cat bond AIG is seeking a source of fully-collateralized reinsurance capacity for U.S. wind risks (hurricanes, tropical storms etc), with the deal structured to provide per-occurrence protection using a parametric index trigger.

Being the first catastrophe bond to utilise a parametric trigger since April 2009, it’s interesting to have a little more detail on how the index is constructed and how transparent, or otherwise, it is for investors to understand.

One of the key’s to a parametric trigger is in transparency, making it simple for both sponsor and investor to appreciate and to understand how the event impact is translated into an index value which can be compared to a trigger point.

In this case, it seems that while not the simplest form of parametric trigger, there has been enough detail provided to investors and to rating agency Fitch, in order to help them understand the parametric index construction to get comfortable with the deal.

Firstly, Fitch explains that the trigger would respond to events on a per occurrence basis, with the underlying trigger factors based on a parametric index. Fitch notes that this means that there will be a basis risk, because the event payout may not be the same as the actual claim experience of AIG.

It’s questionable whether sophisticated ILS investors are particularly bothered about basis risk with parametric triggers, as long as the deal is structured well and comes furnished with sufficient data and transparency to allow them to make a proper assessment of the risks.

The fact that basis risk exists in a parametric cat bond is well understood, but investors are typically allocating capital to parametric deals based on the risk factors and the return, not so much on whether a cedent actually suffers a specific level of loss from a storm.

Basis risk is also likely not so much of an issue to AIG as sponsor either. Being a large, sophisticated insurer, AIG will look at the protection from this transaction very much as a hedge which would provide a source of liquidity after a major event, whether it suffered significant losses or not. AIG has its indemnity reinsurance and cat bonds in-force to provide it with protection for specific losses already.

Basis risk aside, Fitch goes on to explain the mechanics of the parametric trigger and the construction of the index value which would determine whether the Compass Re II cat bond faced a loss or not.

The parametric index trigger is based on the following parameters; latitude and longitude of the storm at landfall, radius of the storm and the maximum sustained wind speed. By bringing these factors together with weighting based on the location of landfall, a parametric index value can be determined.

Fitch explained:

The Windstorm Parameters of latitude and longitude at landfall and radius determine the Parameters X and Y used in calculating the Event Index Value. The Event Index Value is a mathematical formula that multiplies the Parameter X by the Maximum Sustained Wind Speed to the Parameter Y exponent.

If the wind speed and radius are not available at landfall, then the largest of the reading just before and after making landfall will be used.

If the Windstorm intersects multiple Boundaries, the respective Event Index Values will be summed to determine the Event Percentage (or the amount to be paid to the Ceding Insurer). There is a complicated linear interpolation formula, which is easily traceable in a spreadsheet format.

It’s certainly not the most straightforward of parametric triggers, as the event index value is determined largely by where a storm makes landfall. However, that allows AIG to calibrate the trigger to be most responsive in areas where a landfall is expected to cause a greater loss to the insurer, thus helping to reduce basis risk from the sponsors point of view.

Fitch continues:

Noteholders are exposed to principal loss if the Event Index Value exceeds 100.0 and face total principal loss if the Event Index Value reaches 150.0.

Assuming maximum sustained wind speeds of 100 mph and a 100 mile radius, the Notes would be triggered if a wind storm crosses certain parts of New Jersey coastline, but no other areas along the U.S. coastline.

Increasing the wind speed to 130 mph while maintaining the 100 mile radius, the Notes could be vulnerable if landfall occurred to certain parts of the Texas, Florida, Virginia and New York coastline. A category 4 hurricane is defined with wind speeds in excess of 130 mph.

So clearly the Compass Re II 2015-1 catastrophe bond is particularly sensitive to a New Jersey located hurricane or extra tropical storm landfall, likely as a result of the experience with Sandy.

Fitch explains that in terms of historical events, none that it looked at threatened the notes parametric index attachment point.

A repeat of Superstorm Sandy would result in an event index value of 25.275, Hurricane Rita would have been 21.642, while Hurricane Frances (a storm that made two landfalls) was just 3.034. Hurricane Camille in 1969 would have resulted in an Event Index Value of 61.135 using on an assumption that the radius of the storm was 90 miles.

Risk modelling firm AIR Worldwide provided a modelled analysis for the Compass Re II 2015-1 cat bond notes to assist with the rating process, the firm also acts as calculation agent.

AIR simulated 10,000 wind storm events to assess the notes, with its model producing an attachment probability of 2.41% which corresponds to an implied rating of ‘B+’ according to Fitch’s rating methodology.

A stress test using the Warm Sea Surface Temperature Conditioned Catalog had an attachment probability of 2.67%. The expected loss for the single tranche of Class A Compass Re II notes is 1.77%, 1.94% for the WSST catalog.

Being the first catastrophe bond covering U.S. wind or hurricane risks to use a parametric trigger since April 2009, this Compass Re II deal is likely to test investors appetites for parametric hurricane risk. It will be interesting to see just how readily it is received, initial reports from investors we’ve spoken with seem positive.

This catastrophe bond is also the first to be structured by Rewire Holdings LLC, the firm that launched in September 2014, as well as the first to be marketed using its web-based marketplace, Rewireconnect, where investors can buy, sell or trade in ILS and catastrophe risk assets.

With a number of novel factors in this cat bond, as we wrote last week, successful issuance can never be assured.

However, the fact that the deal only has a six month duration, a unique feature of this cat bond, is unlikely to be a particular issue, given that many ILS investors are already used to allocating capital to transactions that only span a single U.S. wind-season, such as cat-in-a-box and certain ILW deals.

As ever, whether this cat bond is successful or not will more likely come down to the pricing offered. Investors appreciate parametric triggers and have experience of transactions spanning less than a year. As long as the premium being offered by AIG (remember this is a zero-coupon note) is sufficient to compensate for a more complicated trigger, the six month risk period and the zero-coupon structure, there is no reason to believe it would fail at issuance.

We will update you as the Compass Re II Ltd. (Series 2015-1) catastrophe bond continues its journey to market. You can read all about this deal and every other cat bond in the Artemis Catastrophe Bond Deal Directory.