1 March 2011 Insurance

Trouble in the outback: reviewing catastrophe losses

The concept of a catastrophe excess-of-loss reinsurance should be a simple one. It usually covers losses occurring during the period of the contract on policies specified and covered by the scope of the reinsurance arising from a form of words similar to “any one occurrence and/or series of occurrences arising from any one event and/or catastrophe”.

But while it should be simple, reinsureds will note that reinsurers have often drafted (and used) clauses that have proved to be precisely the opposite.

The first of these clauses limits the duration of any one catastrophe (generally in the case of flood perils) to losses occurring within a 168-hour period. The reinsured may choose its inception date provided that it is not before the first recorded loss date and also provided that no two periods overlap.

This clause has caused many reinsureds (abetted by their IT systems) to believe they have purchased an aggregate excess-of-loss cover. They have not—they have still purchased, as per the hours clause, a cover occasioned by one catastrophe.

The second well-meaning source of confusion and potential argument is found in the extended expiration clause, which entitles the reinsured to claim on an expiring reinsurance cover (subject again to overlapping provisos) should a loss occurrence commence (for instance) on the last day of the expiring cover.

Claiming on a cover on its last day has advantages. It avoids the necessity of purchasing back-up cover, which could be expensive. In the unlikely event that reinstatement premiums are calculated pro rata as to time as well as amount, there are obvious economies to be made by claiming on the old cover rather than its renewal.

It also ensures the certainty of contract if the placement of the new coverage is incomplete. As an historic reference, the UK windstorms, which took place overnight between December 31, 1975 and January 1, 1976, threw up many disputes that were finally sorted out in the then customary ‘gentleman’s compromise’.

Finally, the loss settlement clause reinforces the legal position that the reinsurance can only respond to losses where there is policy liability to pay. Both the Australian federal and the Queensland state governments show strong signs of applying pressure to insurers to pay flood losses on policies where flood perils were not covered.

Reinsurers, not only in respect of the contractual agreement but also having rated the business in relation to aggregate flood exposures, can rightly point out that they will only respond to losses where there is policy liability. However, policy liability can be distorted after a catastrophe loss where the costs of reinstatement are necessarily increased because of the shortage of the tradesmen and building supplies.

A proposed approach

The definition of any one occurrence in a state the size of Queensland paradoxically should not present a problem because the definition of occurrence relies on the unifying factors of time, place and cause. A reliminary review of the timing and location of floods affecting the state would seem to indicate that there may be several distinct occurrences at Rockhampton, Maryborough, Bundaberg, Condamine/Chinchilla and the Brisbane river basin from Toowoomba to Brisbane.

Any suggestion that all the events are linked in time (and can be aggregated within a 168-hour time period) and place (separated by several hundred miles over different river basins) and cause (the movement of the La Niña current?) is not feasible.

There has also been, as widely reported in the media, yet more flooding in Australia principally in the state of Victoria caused by heavy rainfall between January 12 and 14, 2011. These subsequent floods caused a further assessment of the events in question.

The Australia Bureau of Meteorology (maybe unhelpfully) attributed the severity of the rainfall to the position of the La Niña current, which caused the drawing in of warm tropical air from the north with its high level of humidity to conflict with cooler air from the south, thus causing thunderstorms to form over Victoria and partly over the neighbouring states of South Australia and New South Wales.

The result of rainfall in the region of 300 to 400 mm (almost the same as the annual average) was the worst flooding in the history of the state. The floods devastated 51,700 hectares of pasture and 41,200 hectares of crops. In addition, 6,106 sheep were killed. The Department of Primary Industries has calculated a damage bill of up to A$2 billion.

Likely coverage and losses

Despite the size of the area covered (some 100 km by 100 km), the area is sparsely populated. one of the largest towns affected, Horsham (located in Victoria 300 km from Melbourne on the Melbourne to Adelaide railway), has a population of only 14,125. The insurance bill is likely to be moderate.

In the ordinary course of events, a claim falling on the reinsurance market from catastrophe excess-of-loss contracts would appear to be unlikely. However, what would be the position if a reinsured were to submit a claim aggregating the losses from Victoria with those from Queensland?

The justification for this could be that the rainfall was caused by La Niña and the dates are sufficiently linked in time to allow the losses to be aggregated within the usual 168-hour time period. The benefit of this approach (subject to there being sufficient coverage available) is obvious in that the deductible is consumed entirely by the Queensland event so that all the Victoria losses are recovered from reinsurers.

However, the facts are at variance with this approach.

Firstly, Horsham is more than 1,500 km from Brisbane on a river system that drains south into the Great Australian Bight rather than draining east into the Pacific Ocean. Secondly, the rains causing the floods happened at a different time and place to those causing the Brisbane floods. And finally, the flooding took place later and was not contiguous with that in Queensland.

It is not the same catastrophe linked uniquely by time, cause or location, so cannot be linked with the earlier Queensland floods.

Peter Smith is an associate with GKA Associates. He can be contacted at: Peter.Smith@GKA-Associates.com

Already registered?

Login to your account

To request a FREE 2-week trial subscription, please signup.
NOTE - this can take up to 48hrs to be approved.

Two Weeks Free Trial

For multi-user price options, or to check if your company has an existing subscription that we can add you to for FREE, please email Elliot Field at efield@newtonmedia.co.uk or Adrian Tapping at atapping@newtonmedia.co.uk