Facing the challenge of new threats

Can chemical, biological and nuclear attacks be covered and how can such risks be priced?

David James, senior underwriter for political risk and terrorism at Ascot Underwriting summed it up by saying he would try not to write the risks as they could not be technically priced.

He said: “We don’t feel they can be fitted in easily with the current aggregation model.”

James said pooling might be a good option, and that Pool Re provided cover for this. But other insurers would be prepared to cover it as they were comfortable quantifying the risk. He also questioned how relevant the cover was as Ascot wrote a lot of policies that excluded such risks, demonstrating that policyholders were prepared to buy cover without them.

He added that many clients considered such risks to be a “government peril”.

Kirsten Parker, director of analysis for Exclusive Analysis, said there needed to be more work to determine the real exposure of chemical, biological and radiological attacks and what could be written and what was too heavy a risk.

What were the policy wording considerations?

David Chidwick a partner at law firm Mayer Brown International, said the wordings would need to be clear on how much insurers would have to do to decontaminate an affected site.

Current wordings, he said, did not determine how much insurers should spend on clean-up and a lot of post-event factors would have to be included.

Andrew Coburn, vice president of catastrophe research and director of terrorism research at Risk Management, added that a lot of current modelling was based on assumptions such as whether decontamination was done on site or whether it would have to be shipped elsewhere. This led to uncertainty.

He said: “Clearly in situations like this there would be heavy government involvement. This is an exposure that clients want cover for and, if insurers don’t want to bet their business’s offering, then that falls on the government.”

James said there would be a willingness on the side of insurers to offer this but clients would have to pay exceedingly high premiums for that kind of cover.

“Some big explosive attacks can easily be brought into key modelling, as well as where the attacks are likely to happen

Andrew Coburn Risk Management Solutions

How accurate is risk modelling in assessing terrorist threats? And does risk modelling work?

Coburn said loss modelling was getting better and was quite effective in dealing with conventional weaponry attacks, by measuring factors such as blast radius.

He said: “Some big explosive attacks can easily be brought into key modelling, as well as where the attacks are likely to happen and what the key targets are.”

Coburn added that 48% of attacks focus on the prime city of a country. Modelling can determine how targets change from year to year and monitor how trends change in terms of targets within the city.

Parker said modelling could assist in plotting out the degree of an attack, to analyse how far terrorists got in launching their attack and how likely they were to succeed.

It could also be important in analysing how different terror groups learned from one another.

Ascot’s James said it was important to keep in mind that these were only models and sometimes models broke or were wrong.

He said: “They are based on lots of assumptions and you can sometimes mistakenly believe that models are based on science. The bottom line is that you can use these tools and models and assess risk but really you just don’t know.

“The terrorist is constantly assessing the maximum impact and we have to be aware of that. Models are good, but they are not enough.”

Parker said models were about understanding groups and data and not just about keying in variables.

Coburn added that models could help with information such as how big should the blast zones be. He said: “They are not predictive, but they are about relativism.”

James said what it really came down to was that underwriters wanted a model that could tell them in the event of an attack, how much they had lost. If the model did not do that, he added, than it was not worth having.

“The terrorist is constantly assessing the maximum impact and we have to be aware of that. Models are good, but they are not enough

David James Ascot Underwriting

How real is the threat of cyber and other forms of ‘soft’ terrorism?

Parker said there had been a lot of evidence gathered on terrorist plots to target IT systems, but she said it was important to consider how many groups had the capabilities. Foiled attacks had shown to be planned against the support services of companies and not the targets one might have assumed.

Coburn said the US military experience thousands of probe attacks against their IT systems a month, usually from places such as China or the hacker communities.

Very few of those attacks, he added, had been destructive and they just delay the internal service.

Nonetheless, Coburn said cyber attacks were listed within the top three concerns of risk managers.

With extra capacity coming into the market, will the aggregate limit be removed?

James said he would not be keen to see the cap removed but it was something that Ascot would continue to monitor. Coburn added that a limit was needed to effectively manage the risk.

What value are clients getting from doing risk surveys and is it worth it?

The consensus seemed to be that underwriters were more willing to write business in areas they wouldn’t have before and with the current soft market conditions came real opportunity.

Coburn said risk surveys could be used to help underwriters find new opportunities and that there was increasing demand for international analysis.

James agreed that the risk surveys might help underwriters to consider new risks, particularly with emerging markets.