The full details of weather events are not known for years, by insurers or anyone else who studies weather professionally. For example, the National Oceanic & Atmospheric Administration revealed two years ago that 1992's Hurricane Andrew was of greater intensity than had been previously imagined. The storm was upgraded from Category 4 to Category 5 after hurricane experts concluded that Andrew's winds roared in at 165 mph--20 mph faster than earlier estimated.
The results of the continuing Hurricane Katrina analysis initiative being undertaken by the insurance industry change almost with every passing day. That can render definitive statements on such events of fleeting value, even years after the event.
It is clearly too soon to form many conclusions about the insurance and reinsurance effects of hurricanes Katrina and Rita, but events in the Southern states have kicked up some issues that may spawn their own winds of change in the insurance industry.
For instance, we know that the traditional ratio of commercial-to-residential damage changed for Katrina. Was that a one-off, or evidence of a broader shift in the nature of coverage? How has the growing hedge-fund interest in reinsurance stood up to Katrina, what by most accounts may prove to be a $60 billion insured loss, and the subsequent damage caused by Rita? Will the unending images of New Orleans as a submarine city encourage wider insurance coverage? And just how convoluted will the claims end of the process prove to be?
Some answers were apparent within a few days of the shock wearing off. As their estimated share of the Katrina losses were reported, and then updated, by the insurance and reinsurance companies it was clear that with a loss of such magnitude, many claims will reach or exceed maximum policy limits.
If the $40 billion to $60 billion loss estimate holds for very long (it is almost bound to rise as years go by), the reinsurance industry would be able to absorb it without much difficulty (except in certain individual cases), and perhaps even without exerting much upward pressure on premiums. As the ever-quotable Jim Bryce, chairman of IPC Re Holdings, said soon after Katrina hit, paying catastrophe loss claims is "what we do."
The Bermuda reinsurance industry share of a $40 billion to $60 billion loss from Katrina would fall to between $7 billion and $10 billion, net of everything. If one or two of the companies--names are being bandied about, but none is confirmed--prove to have much deeper coverage in the South than had hitherto been expected, losses would be nearer the higher end of that range. Out of a $100 billion estimated capital base, not to underestimate the appalling human cost of Katrina, that's business as usual.
Given the probable damage to oil platforms and refineries in the Gulf, the financial well-being of OIL Insurance Ltd. was an early concern. The company promptly reported that it has a per-event aggregation limit of $1 billion, and a capital base of $4 billion. As a mutual-like company, OIL will recover the loss from members over a five-year period.
An educated guess suggests that most of the Bermuda companies would have been able to meet their share of Katrina losses out of current year earnings in an otherwise quiet quarter. The problem is that even before Rita the third quarter of 2005 has been anything but quiet. Several other industry loss events in the quarter included five aviation losses, Atlantic hurricanes Dennis and Emily, two typhoons in Japan and China, and flooding in India and Central Europe.
The combination of Katrina and Rita with all these events, plus whatever else the second half of the hurricane season and the first half of the winter storm season in Europe throws at the reinsurers, means that property rates for 2006 can only increase. Indications are that some less well-capitalized U.S. companies might feel the heat and exit the market, but that is how reinsurance works with less well-capitalized U.S. companies. Will we see a tranche of new insurance capital forming in Katrina's wake? Not on the evidence so far. Will the same hold true for Rita? It's too soon to say.
Which brings us to ways in which Katrina is out of the ordinary, in reinsurance terms.
Business interruption covers loss of profits in the period before a company can be put back on its feet. With New Orleans essentially abandoned to dry out for a period of months before rebuilding starts, business interruption claims are expected to reach 50 percent of the total insured loss. They usually run nearer to 30 percent. The obvious conclusion is that claims will reach policy limits and trip the excess layers. business interruption rates and terms for next year will definitely be harder.
The change in the commercial/residential ratio would probably only apply in circumstances where an entire city is effectively destroyed. It is hard to imagine a scenario other than New Orleans or perhaps a part of the Netherlands in which that would be the case. To make Miami, for example, a completely uninhabitable zone other than by tidal wave would require a storm of unimaginable strength. Anything is possible. Category 5 storms cause "total destruction" if they score a direct hit.
The position of the hedge funds with regard to Katrina is harder to divine because privately held ventures are not required to provide sufficient public information to make detailed analysis possible. A chorus was heard soon after Katrina struck, to the effect that the hedge funds would not be deterred by the storm. Logically, given the nature of the risks that such funds are believed to be underwriting, that position does not sound entirely tenable.
Hedge funds are reportedly writing specific reinsurance contracts that kick in at high excess levels; other contracts are very time- or geography-specific. Unless hedge fund reinsurance managers have enjoyed an unlikely stroke of luck, it seems probable that one or more of them will have suffered a significant financial loss. Whether that information will ever be made public is a separate question.
MORE FLOOD COVERAGE?
An issue swirling through the reinsurance market in the information hiatus before loss estimates were released was whether the unremitting news focus on the Gulf Coast would result in more people buying insurance coverage, especially flood coverage. The general view, delivered with a shrug of the shoulders and a furrowing of the brow, was that it would not. Businesses tend to have coverage; it is often individuals that do not. The villain is the oldest self-insurance argument in the world: "It won't happen to me."
Damage caused by Hurricane Katrina was followed two days later by damage caused by what history will recall as the Great New Orleans Flood, which washed throughout one of a very few U.S. cities to have retained a unique personality. Attributing the right share of damage to the correct event will prove a difficult and time-consuming exercise for claims analysts.
If the estimates are to be believed, Katrina will cost the insurance industry about one-quarter of the total economic loss. That figure is low (50 percent is traditional), in large part because of the extent of flood damage.
So the answer to the question, "Might Katrina cause any significant long-term change in how reinsurers underwrite and look at this kind of catastrophe?" appears to be "no." Rita may yet prove to be a different story.
The modelers' knowledge base will be enhanced. A singular event, the loss of a modern city, can be added to the model possibilities. But every storm brings its own special circumstances, and the double catastrophe of a storm followed by the failure of the pumping system of a city, much of which is built below sea level, ought to be a rare event indeed, of interest more for its singularities than its greater lessons.
ROGER CROMBIE is a writer based in Bermuda. He is a regular columnist for Risk & Insurance®.
October 15, 2005
Copyright 2005© LRP Publications