1 October 2011Re/insurance

Terrorism coverage: uncertainty is the only constant

The shadow of September 11 still looms large in terrorism circles, but terrorism reinsurance pricing has moved on significantly since those tragic events 10 years ago. The post-September 11 period can be seen as a high point for terrorism pricing, with an extended softening in the intervening years leading to the essentially flat pricing environment in which the line finds itself today, said Kristopher Lynn, assistant vice-president, underwriting at Tokio Millennium Re. Price declines have been around 5 to 10 percent year-on-year, he said, with some variation by geography—“with European pools tending to be rathersteadier than those in the US”—but generally pricing has been on an extended downward trend in Europe and the US.

About the only area where there has been an “uptick” in pricing has been on “worldwide covers, including those typically underwritten by Lloyd’s syndicates”, Lynn said. Ongoing events throughout the world have helped to prompt localised price increases on such books of business, he said, although he added that it’s “worth noting that these types of covers include broader coverages such as strikes, riots and civil commotion, which may well be driving pricing more than straight terror”.

But it seems that pricing on terror reinsurance has reached something of a nadir that in many ways reflects the wider reinsurance market. As Mirek Wieczorek, senior vice-president, specialty and casualty lines at Tokio Millennium Re, made clear “after years of rate decreases—and this is in general, not only on terrorism lines—pricing has flattened off. Multiple capital eroding events over the last 18 months have not proved big enough to turn the market, but they have helped to instil greater pricing discipline in the market”. This will translate into the terrorism space, he said, adding that a firming up of rates was necessary if companies, institutions and governments hoped to continue to enjoy the continued support of reinsurers.

"The riots in Thailand are a good example-- we observed rate increases on the back of these events, but only locally."

Should pricing continue to dip, it would seem that the appetite to continue writing the line—even considering the ‘diversifying effect’ of specialty lines such as terrorism—will wane, with all the implications that this would have should another major terrorism event occur. As Lynn indicated, “a lot of re/insurers have short memories and as new markets come online they tend to compete on price. Terrorism is by its very nature a diversifying peril, and as such there is a danger that firms write the line at a lower price because it helps diversify their book, with such an approach playing into the downward pricing dynamic”. However, it would appear that the current pricing environment is unlikely to encourage an uptick in interest, particularly considering the specialist nature of the perils underwritten.

Pricing: at the mercy of radicals

Addressing the major drivers of pricing, Wieczorek outlined three downward and three upward pressures that determine the heading of terrorism reinsurance.

Downward drivers

A lack of frequent large-scale attacks “We feel that the market tends not to price near misses or attacks in other geographies, rather its approach to pricing is more localised,” Wieczorek said. “An example would be the car bombing attempt in Times Square, which had no discernible impact on pricing, but had something happened, then the pricing environment would have been significantly different.”

Supply and overcapacity “This is perhaps the biggest downward pressure on price,” Wieczorek said, “one that is driven by the diversifying nature of terrorism on the book of business.” With Bermuda reinsurers writing predominantly property cat books of business, there may well always be a temptation to explore specialty books such as terrorism at a competitive price.

Demand “Demand for terrorism reinsurance is not growing, but remains relatively stable,” Wieczorek said. Only a major terrorist event looks likely to change this dynamic.

Upward drivers

Events “Events, when they occur, have a definite impact upon pricing locally,” Wieczorek said. “The riots in Thailand are a good example—we observed rate increases on the back of these events, but only locally. And that is the issue: the Thai riots are indicative that people do not fully appreciate the full implications of events when considering worldwide accounts,” he said. Rather, events tend to alter pricing only in their immediate geography.

The terrorism risk insurance act “If the next administration were to scale back the current federal terrorism backstop, which is due to expire in 2014, there will likely be a rise in potential risk, which would inevitably be reflected in pricing,” Wieczorek said. “However, we feel some kind of similar programme will be put in place,” he added, and the signing of the Terrorism Risk Insurance Program Reauthorization Act (TRIPRA) in 2007, which extends the government backstop until 2014, suggests that US government coverage will remain in place for the foreseeable future.

Market-changing events “A significant market-turning event still has the potential to drive pricing upwards,” Wieczorek said. “The terrorism market will inevitably change as terrorism threats evolve,” he said, adding that it is very difficult to predict what new threats will emerge.“If you look at the last 12 months—unrest in the Middle East, the attacks in Norway and the killing of Osama Bin Laden—they simply demonstrate that the situation can change rapidly.”

Uniquely challenging

Turning to the unique nature of writing a terrorism book of business, Wieczorek outlined three specific challenges underwriters face.

The frequency of events While Tokio Millennium Re uses vendor models in order to “track aggregation” and provide its team with a “further view on pricing, the biggest assumption and challenge is related to frequency,” Wieczorek said. “When modelling terror events, we have the ability to measure the impact of various sizes and severities of events, and even locations, but what we can’t model for is frequency. We feel that when it comes to vendor models, it is rather more difficult than compared to, say, hurricane models. When you estimate wind events it is significantly more scientific than when you project terror events.” Judging frequency is therefore a unique challenge.

Multiple events “There is always the chance of subsequent or simultaneous attacks in one, or even more than one, country. So modelling multiple events is highly uncertain,” Wieczorek said. “This means that we spend a lot of time testing modelling assumptions in order to get a handle on our exposures.”

A constantly evolving risk landscape “When modelling risk we are often focusing on yesterday’s terrorism problems, rather than tomorrow’s. And this makes unusual scenarios difficult to imagine before they occur. As such we are often fighting yesterday’s wars,” Wieczorek said, leaving reinsurers exposed to potential emerging threats that could blindside the industry.

Terrorism plus

Turning to coverage, Lynn said that Tokio Millennium Re’s terrorism book of business is predominantly made up of European terrorism pools, with some US and worldwide coverages complementing its preponderance of European coverage. “A typical attack would be a bombing in a major city, such as the London bus and underground attack in 2005,” he said, but the team at Tokio Millennium also writes chemical, biological, nuclear and radiological (CBNR) coverage and more packaged risks that tend to blur terrorism with other civil disobedience. It is “within Lloyd’s that you tend to see a packaging of risks—you get terror, but you also get riots, civil commotion, strikes, and to some degree marine war or aviation war”, Lynn said, “with the book typically extending across several classes”.

“A perfect example of this blurring is the Thai riots, which will be in the courts for years to come determining whether it was terrorism, civil commotion or civil insurrection,” Lynn said. “The outcome in such disputes will very much depend upon the laws of the individual country. Typically, for worldwide coverages you could be underwriting risks for an entire hotel chain. As such you would be picking up exposures across the globe, with some of those countries being far more unstable from a terrorism point of view than others,” he said, while some would include threats outside the usual gamut of ‘terrorism’.

Due to the broad scope of some of the books of business— particularly those coming out of Lloyd’s—and the unique nature of the risks underwritten, specialist third party institutions are key to understanding and pricing terror risk. Lynn said that as a reinsurer Tokio Millennium looks to its cedants to fully understand and assess their risk exposures. “What I like to see is that insurers are using third party institutions such as Exclusive Analysis, Control Risks or Red 24, which are institutions that collect information on security and politics worldwide and produce reports for insurers,” he said. “Before writing a risk in, say, Bangkok, they need to take this information and ask, is this a risk we really want to write against, or should we instead be writing somewhere else?” Much as catastrophe models inform those writing wind and earthquake books of business, such reports are an invaluable aid to re/insurers in understanding and pricing terrorism risk.

Government backstops

Looking back to the tragic events of 10 years ago, Lynn said that September 11 proved a game-changer for terrorism re/insurance, prompting governments “to participate directly in the overall solution”. Washington’s TRIPRA and the establishment of similar intergovernmental risk pools in Europe were a response to exactly this need. It was understood that the “involvement of governments is the only way to ensure pricing stability” as few re/insurers could take on the extreme tail risk that a huge terrorism event would generate.

Government risk pools were recognised as the only institutions that can truly “step up to the plate” should an extreme event occur, he said, with re/insurers able to shoulder the burden of smaller, more frequent attacks, and a proportion of cost should a major terror incident occur. Terrorism pools look set to be a continuing component of the landscape for a time to come, said Lynn, providing valuable additional capacity to the market.

With pricing having been on a downward trajectory in recent years, Lynn said that it has become even more important to have “government pools as a backstop for capacity”. However, while an invaluable aid, support for insurers from the likes of TRIPRA “doesn’t always translate into the reinsurance arena,” he added. Rather, it means that cedants “only buy reinsurance up to their trip-wire retention”. This creates a stable environment for insurers, but constrains ceded premiums for reinsurers.

Nevertheless, there may be opportunities for reinsurers to support programmes such as TRIPRA. As Wieczorek made clear, “if we were to have another major event in the US, I think that TRIPRA could be looking to the private market for support and capacity. Right now TRIPRA is not buying any reinsurance, but if the backstop was to face significant uptake as a result of a major event, then probably private market involvement would increase and be involved—or at least be explored—in the next version of TRIPRA”.

An uncertain future

Looking ahead to likely opportunities in terrorism reinsurance, Lynn said that barring “a market-changing event, we don’t actually see a lot of opportunities going forward. From the threat perspective there will always be terrorism and its nature will always develop, but there are no major opportunities out there at present”. Instead—and barring a major event—pricing looks set to be flat, with only localised spikes in those geographies directly affected by terror events. Much, it would seem, has changed since the dark days following September 11.