Written by Helen Yates
From floods and hailstorms in Central Europe and tornadoes, floods and wildfires in the US to intense cyclone activity in the Pacific Basin, 2013 brought as diverse a range of nat cats. Yet, despite forecasts for an active year in the North Atlantic, this year’s hurricane season was remarkably muted.
“If we look back to the Central European floods, the hail storm in Germany, bushfires in the US, floods in Canada and Colorado and that there’s been twice the cyclonic activity in the western Pacific than on average – there have been a lot of catastrophic events in 2013 although not one that moved the market,” says David Lightfoot, head of GC Analytics. “So in my mind it reinforces the benefits of diversification and it does show that an active cat risk management strategy is appropriate.”
2012 was a more “typical” year in the sense that the US dominated from an insured loss perspective, and the biggest loss – Hurricane Sandy – came in the second half of the year. By contrast, the largest loss in 2013 came from Europe in the first half of the year, with the Central European floods in June expected to generate claims of US$5.3bn (according to Aon Benfield) and hail storms in Germany in July generating a claims burden of approximately US$2bn to US$2.7bn (according to Swiss Re).
“2013 has been quite an interesting year in the sense there was no significant event in the Atlantic, so a lot of attention has been given to events outside the US,” notes Dr Milan Simic, managing director of AIR Worldwide. “A lot of attention was focused on Germany with hail and flooding there. Another trend was the very high activity in the Pacific basin – which is the most active basin in the world – but we had a lot of landfalling events affecting a large number of countries.”
While there was little activity from North Atlantic hurricanes, extreme US weather was once again a feature during 2013. On 20 May, an estimated EF5 intensity tornado near Moore, Oklahoma was responsible for 25 fatalities and causing insured losses ranging between US$2bn and US$5bn, according to Eqecat. And in November – a month not usually associated with tornadoes – there was an outbreak of intense thunderstorms, with high winds, hail and dozens of tornadoes across the Midwest.
Floods were also a big source of losses in North America in 2013, with historic flash flooding in Colorado killing nine people in September. This year’s flood events have highlighted the fact it remains an underinsured peril, even in Europe and the US, with few of the estimated US$2bn in economic damage from the Colorado floods expected to be covered by the US National Flood Insurance Program.
Heavy rain and flooding in Alberta, British Columbia, in June generated insured losses of around US$2bn, the highest insured loss from a single event ever record in Canada. Strong thunderstorms in the greater Toronto metropolitan region, resulted in Canada’s second billion-dollar natural disaster event of 2013 in July.
In the latter half of October, Europe was pounded by Windstorm Christian with a billion dollar insurance toll, according to Aon Benfield. The heaviest damage was sustained in the UK, France, Belgium, the Netherlands and Scandinavia. At this early stage of the European winter storm season it is the costliest windstorm since Xynthia in February 2010.
One theme that continues to dominate with losses in emerging markets is the large gap between insurance losses and economic losses, a gap that is widening according to many studies. Despite the devastation and high death toll it caused in the Philippines, Super Typhoon Haiyan is unlikely to be a significant event from an insurance perspective, with AIR Worldwide predicting claims between US$300m and US$700m versus economic losses of US$6.5bn to US$14.5bn.
The Philippines is no stranger to typhoons and other catastrophes, with exposure to earthquakes and volcanic eruptions among other perils. However, it has very low levels of insurance penetration (although there are reports the government will soon make earthquake insurance compulsory). At present the insurance penetration is just 0.5%, according to estimates from Swiss Re.
“If the storm track had gone to the north through Manila (which has a population of 20.9 million) the insurance losses would have been higher and also if it had impacted more than one industrial park the insurance loss would likely be higher,” says Lightfoot. “With the Thailand floods in 2011 the majority of losses came from the industrial parks. Only one industrial park appears to have been impacted by Haiyan, which will probably end of contributing a considerable amount to the insurance loss, but as such is not drive a large loss for the industry.”
“There is capital to increase insurance penetration in countries like the Philippines,” he continues. “One of the contributors to increasing penetration would be increased ability to quantify risk in these emerging markets that have low penetration. But as important is the need to provide and find solutions that fit the needs of the people and companies that would benefit from insurance. It will take the industry to be creative and work with interested parties to find these solutions and there is appetite and strong interest in the industry to do this.”
Haiyan is the strongest storm to make landfall in recorded history, with wind speeds of up to 195mph when it struck the Philippines on 8th November. It was also the fourth strongest cyclone ever recorded, according to the Joint Typhoon Warning Centre in Hawaii. The storm claimed over 4,000 lives when it made landfall in the city of Guiuan (population 47,000) before tracking through Tacloban City (population 220,000).
“The approximate track of the cyclone was known for a good number of days beforehand but what was more difficult to understand, and responsible for the high number of deaths and casualties in this case, was the intensity – and that’s something that’s relatively more difficult to forecast,” explains Simic. “Also the extent of the storm surge wasn’t that well understood and there was little or no warning to the local population to evacuate.”
“That part of the world has relatively few anemometers so it was difficult for us to get reliable estimates of wind speeds in the wider area,” he continues. “Another thing that contributed to the strength of the storm is the fact the Philippines is a group of islands. When a hurricane makes landfall in the US there is typically a rapid reduction in strength as the storm moves over land, and this wasn’t seen in the Philippines because of the nature of the terrain.”
Climate change and catastrophes
As is typically the case in the aftermath of a major weather-related catastrophe, questions were raised about the intensity of Haiyan and influence of climate change. UN secretary general Ban Ki-moon said the Philippines super typhoon was an “urgent warning” and an example of “how climate change is affecting all of us on earth”. However, in its fifth impact assessment the Intergovernmental Panel on Climate Change (IPCC) found only “low confidence” that intense tropical cyclone activity has increased measurably since 1950.
“Many climate scientists predict that we can expect more intense events in the future, but very rarely is a definite horizon given for that,” notes Simic. “On the other hand scientists are very careful to mention that no single event can be attributed to climate change, but then we have a situation that politicians, when they see the destruction around the world, say ‘something must be happening’.”
The high levels of storm activity in the Pacific basin in 2013 are more likely to be linked to a short-term climatic cycle than long-term anthropogenic climate change, thinks Lightfoot. He notes the Western Pacific cyclonic activity was around twice the average in 2013. “There is good scientific research out there that links cyclonic activity to the La Nina cycle and that is being considered as companies set their overall risk management strategy,” he says.
“There’s the report by the IPCC with respect to climatic change but that report is more about long-term trends,” he continues. “For example it predicts there will likely be higher intensity storms in the future but they will occur less frequently and there will be a gradual rise in sea levels. Those are long-term trends and property insurance is a generally a short term financial instrument, and considering long-term trends when modelling for a short-term instrument is challenging.”
Sea level rise is significant from a storm surge perspective, with some experts questioning whether the devastating 13.88 foot surge from last year’s Hurricane Sandy was exacerbated by sea level rise. The storm’s timing was such that its strongest winds and peak surge arrived right as an astronomical high tide was also peaking.
To prepare for the impact of an extreme weather event, AIR has introduced the concept of extreme disaster scenarios (EDSs) as a tool for determining large loss potential. “These take the physical reality of our models to the next level,” explains Simic. “We vary the parameters to make them more extreme events over very high concentrations of insured risk, [for example] the path of a Category 4 cyclone going right over Oahu, Hawaii, an event that is in the order of US$70bn.”
“That EDS principle can be extended to any part of the world,” he continues. “You could look at a typhoon in the Philippines but affecting Manila or other parts of the Philippines that are more densely populated to see the kind of extreme events that could happen.”
2011 was the year that drew the insurance industry’s attention to the potential for sizable losses in so-called “cold spots”. These are parts of the world not traditionally associated with large cat claims. In the aftermath of the second most expensive catastrophe year on record for insurers (with total nat cat claims in the region of US$110bn) and two-thirds of the losses coming from Asia Pacific, a lot more attention was focused on this region.
Simic thinks that is one of the reasons why there is so much focus on this year’s Asia Pacific catastrophes. “There’s been much more awareness and the industry seems better prepared for the next Thailand floods,” he says.
Some of the countries receiving more attention because of their catastrophe exposures and increasing insurance penetration are Thailand, China and Malaysia. A major earthquake affecting Beijing or a super typhoon impacting Shanghai are among the list of scenarios re/insurers consider when stress testing their portfolios.
Rapid urbanisation and the creation of megacities (cities with more than 10 million inhabitants) are also increasing the exposure from a risk and insurance perspective. “If you look at pictures of Shanghai 50 years ago it’s unrecognisable,” says Simic. “The growth in exposure, particularly in Asia, has been tremendous. That coupled with better understanding of events themselves can give us a better idea of the range of losses we’re potentially facing and the EDSs can push it to the bound of possible losses.”
For insurers and reinsurers there was a realisation following the Thai floods that a better understanding of perils and catastrophe exposure was required in some emerging markets.
“Something that’s happened at an increasing rate over the last few years is academics and other institutions becoming much more engaged in this space so there’s a growing list of models and resources that are providing good information in these areas that aren’t covered by the main vendors,” says Guy Carpenter’s Lightfoot.
“That said, certainly there are still non-modelled perils,” he continues. “An appropriate risk management practice is to look at a scenario or RDS [realistic disaster scenario] and see how a significant event for that peril in that territory would be covered by the company’s risk management strategy, including its catastrophe reinsurance programme.”
Download this article as a PDF
Contact the editor