Every so often on Artemis we write about hail and the sizeable insurance losses severe hailstorms can inflict particularly in the U.S. Another example of this has been highlighted after a day of hail storms in Texas on the 13th June caused severe losses to auto and property lines of business. Hail is an interesting peril as it is one that a number of years ago could be transferred to investors through the use of a coupon paying re/insurance product which in some characteristics was similar to a catastrophe bond.
Back in April, Texas suffered a spate of tornadoes which resulted in an estimated insured loss total of $400m for the damage incurred in the state. Now a storm on the 13th June which resulted more in hail than wind damage is thought likely to cause a greater level of insured losses from just that single days hail. PropertyCasualty360 reports here that insurer State Farm have already received 3,800 auto claims and they expect that number to rise, and it’s not just the auto damage to consider, there has been considerable damage to dwelling roofs and structures as well.
The coupon paying re/insurance product we referred to above was the WinCat Coupon which was sold by Swiss insurer Winterthur in the mid to late 1990′s. It was designed to provide cover for Winterthurs book of motor insurance in case of widespread hail damage using a product which paid out to the insurer should over 6,000 cars be damaged by hail or wind on a single day from a qualifying storm(wind speeds of 75km/h and above), in this respect it used a type of trigger akin to a cat bond but not linked to indemnity itself, rather the incidence of a specific type of claim.
The coupon took the form of a subordinated bond which Winterthur sold to investors who were willing to bear a portion of the auto hail risk for Winterthur. The three-year bonds paid an annual interest of 2.25% to investors, but if the trigger event occurred during one of the years then no coupon would be paid. In this way it differs from a cat bond as there was no total loss event, rather the investors just lost interest they were due. The bonds converted at maturity into a number of Winterthur shares, so in return for assuming some of Winterthurs auto risk investors paid a premium for shares but that premium was returned in the form of coupon payments where no triggering events occurred.
We find these a ground breaking product for its time and one which received considerable interest. The idea was that investors would love to buy the coupons as they offered an uncorrelated investment opportunity and a way to diversify their portfolios. With investors seeking similar opportunities today we wonder whether a similar product could be constructed in the current market? The fact that there is no total loss to bear, and simply a loss of coupon payments, would surely make the prospect attractive. If not for hail and auto damage, it seems that it could be possible for liability risks to be transferred using a coupon paying product such as this (perhaps medical device liability claims). The original WinCat Coupon disappeared from the market and there were discussions about issues of moral hazard in identifying what was a hail or storm induced auto claim and what wasn’t. It is likely that a forward thinking insurer could remove the moral hazard issue by choosing a more binary trigger associated with whether a claim occurred or not, rather than the cause, for many lines of business. In some ways the subordinated bonds issued by Winterthur served as a form of contingent capital facility.
Back to hail, it’s clear that there is a need to reinsure against hail damage effectively and the numbers below on last years active U.S. storm season show the size of the issue for insurers like State Farm. In the first half of 2011, which as our readers will be aware was one of the most severe periods of tornado and severe thunderstorm activity on record, State Farm recorded 40,345 auto claims caused be Texas catastrophe activity. That number was up from 4,561 claims in the same period of 2010. Now 2012 is not looking like being anywhere near as severe but seeing claims numbers of that size clearly shows the need for hail to be effectively reinsured. Could we see the return of a hail coupon type product in the future? It’s possible, but for the moment it’s more likely that hail risks would be included in a cat bond or traditional reinsurance policy, until someone tries to launch a new innovative approach to transferring hail risks to the capital markets.
If you’re interested in the WinCat Coupon product we highly recommend you read this paper which discusses their pricing: Estimating the Value of the WinCAT Coupons of the Winterthur Insurance Convertible Bond by Uwe Schmock.
Update: Reuters have reported that local trade group the Southwestern Insurance Information Service are estimating that insured losses from this hail storm event could reach as much as $2 billion, based on the volume of claims they are seeing. If their estimate of a $1.5 billion to $2 billion insured loss proves accurate this would become a very major event and one of the highest insured loss weather events in U.S. history.
If the loss were of this magnitude it could threaten any severe thunderstorm exposed catastrophe bonds as well (of which there are a number in the market). Any aggregate, industry loss based cat bonds would likely see an event of this size contributing to the annual loss amount, it’s less likely that an indemnity cat bond would suffer, whether it would trigger of course is another matter. We believe at this stage that cat bonds will likely be safe but we’ll update you if we hear anymore.