Too much reliance was placed on New Zealand’s own perception of its hazards. A better understanding of risk differentiation is imperative and international reinsurers must put up a stronger challenge to any assumptions being made

Reinsurers have taken a severe beating on the catastrophe front in the first quarter of 2011. Following months of flooding and cyclonic activity in Australia, and severe winter storms in the USA and Europe, an earthquake in New Zealand was not what the market ordered. This event also directly preceded the devastating March earthquake in Japan.

On 22 February, a 6.3 magnitude earthquake struck six miles south-east of Christchurch – the country’s second most populous city. At least a third of the city was reduced to rubble. Catastrophe modelling company AIR Worldwide has reported that insured losses could top NZ$11.5bn ($8.42bn). At the time of going to press the death toll stood at 166, with 118 people missing.

But rather than being one blow too far, analysts believe the Christchurch earthquake – the second to affect the city within six months – could give reinsurers valuable lessons in risk differentiation.

Loss amplification is a major concern for reinsurers. RMS chief researcher Robert Muir-Wood says: “The issue is being exacerbated because Christchurch is built on a swamp.

The EQC [Earthquake Commission, a government-owned entity that provides cover for natural disasters] promises to pay people for damage to the ground underneath their property, which is highly unusual for earthquake insurance.”

According to Muir-Wood, the reinsurance industry will bear the brunt of the commission’s unusual stance. “The EQC is going to go through its policy limits in a lot of cases, and more of the losses will be passed onto private insurers who cover the excess. There is going to be a different proportionality of loss passing to reinsurers between the EQC and the rest of the insurance market.”

Muir-Wood says there have been suggestions that people will be banned from rebuilding in some parts of the city, but he says this will increase claims made against the reinsurance industry.

“There is an estimate that about one-sixth of the population has already left the city and moved elsewhere. The EQC policy doesn’t cover alternative living arrangements, so private insurers may be encouraged to pay for people’s expenses if they have had to move out of their property.”

Better understanding of risk differentiation in New Zealand is now imperative for underwriters. EQECAT senior vice-president Tom Larson says there is a false economy in relying too heavily on modern building codes when classifying overall risk.

“The building code’s focus is to prevent loss of life, not loss of property. In Christchurch, they had a lot of vulnerable buildings and not all buildings were up to modern building codes. A perfect example of this was Christchurch cathedral, which was one of the first to fall.”

Larson also says the industry should be concerned about the use of unsuitable building materials in terms of assessing risk. “In a pre-earthquake survey, about half of the buildings in Christchurch’s central business district were unreinforced masonry. About 60% of those are now destroyed.”

Muir-Wood agrees with Larson but adds there has been too much reliance on New Zealand’s own perception of its hazards.

He says: “When something is internationally reinsured, there needs to be a greater challenge to the assumptions being made by local scientific surveys and where they get their understanding of seismotectonics.”

Earthquake modelling is undeniably complicated, but vital if the reinsurance industry is to profitably classify risk in the region.

Larson says: “Differentiation is key when working with underwriters – you find a model that helps them understand good risk from bad risk and to price them accordingly.”

Muir-Wood says RMS has been conducting experimental work into providing better models for combinations of earthquakes, as opposed to single events. “We are aware there is a huge effect if a reinsurer is hit by two earthquake claims in the same contract period.”

Better modelling will also reduce confusion when differentiating an aftershock from a separate event in contracts – a move sure to please reinsurers.

“A lot of people assume that aftershocks don’t produce as much damage as the initial earthquake, but aftershocks are often bigger than the primary quake. For a reinsurer, it may or may not be in the same contract year and can cause extra nuances because of some of the terms of the contract,” Muir-Wood says.

Many in the industry will see recent events in Christchurch as a welcome wake-up call. “For reinsurers, it is a poignant reminder that risk differentiation remains the focus for underwriting. These attributes that we are seeing demonstrate that we have to remain vigilant,” Larson says. GR