Catastrophe risk, climate change and the convergence market
Reinsurers and insurance-linked securities (ILS) investors have taken some big hits from natural catastrophes in recent years, causing many of them to question how far climate change has fundamentally altered the rules of the game they’re in.
This year alone, Hurricane Ida in Louisiana and the US northeast, flooding in Europe, winter storm Uri in Texas and wildfires in California heaped loss on to loss. Current estimates peg industry losses at about $80bn for 2021, with a good chance it will become the fifth year in a row of above-average losses.
Aon’s Real-Time Climate Change report reflects the general sentiment, describing 2021’s flood events as “highly anomalous,” wildfire as having “taken a dramatic and costly turn in the past decade,” and Ida as “the latest tropical cyclone event to show behavioral change prior to making landfall and after coming ashore,” (it intensified rapidly up to the point of landfall and then slowly weakened as it moved inland).
The current run of significant loss years started when Hurricanes Harvey, Irma and Maria in 2017 contributed $97bn to total insured weather losses of around $146bn, driving the highest loss year on record since 1970. Since then, figures from Swiss Re Institute show the 10-year moving average for insured losses rising steadily year on year after falling earlier in the 2010s, moving from $72.4bn in 2017 up to $80.8bn last year.
This year is looking likely to continue that average’s upward trend.
Returns from ILS have been subdued-to-negative during this period, with the Eurekahedge ILS Advisers ILS index as of September 2021 still 5% below where it was in January 2017. As to reinsurers, return on equity has fallen below cost-of-capital and shown volatility in recent years, with the Aon Reinsurer Aggregate peer group showing returns on equity (ROE) turning downward in 2020, with ROE of 2.3% for the year, though it bounced back in H1 2021 to 13.4%.
Although 2020 brought fresh (re)insurance capital into the market for post-pandemic start-ups, ILS capital has only just started rebuilding after falling in recent years – and this year’s losses have made for a more challenging fundraising environment.
For cat risk analysts, the challenge is how to splice out various factors influencing the size and direction of travel of loss trends over the past five years to hone in on the contribution from climate change to support them in pricing the risk better.
In one version, the recent run of poor performing years is just the unlucky phase in a cycle of weather events that will at some point revert to a long-term, lower historical average.
Climate effects are measurable only over the long-term, after all – and the period from 2009 to 2016 was remarkably benign for insured natural catastrophe losses.
Kevin O’Donnell, president and CEO at Renaissance Re said on the company’s Q3 2021 earnings call: “It's important to note that while storms are increasing in both frequency and severity, some of those changes are cyclical in nature and not tied to climate change.
“In all likelihood, the recent clustering of weather events is more attributable to statistical fluctuation in arrival rates than the influence of climate change. In the same way that the 10-year period prior to Hurricane Irma in 2017 was a statistical outlier, due to the absence of US landfalling hurricanes, the 5-year period since then is a similar outlier for its heightened activity,” O’Donnell said.
Meanwhile, Munich Re CFO Christoph Jurecka, also speaking on his company’s earnings call for Q3 2021, was “still comfortable” with the firm’s catastrophe loss ratio at 8%, where it has been since 2016.
“Cat business is a volatile business, but it is a profitable business long-term,” Jurecka said.
However, catastrophe risk analysts generally don’t tend to argue that losses from natural perils are set to revert to a long-term historical average.
Broker Howden, in its Climate in Peril report, said: “Any expectation that loss experience will revert to the old normal is unrealistic. The past is no longer a guide to the future for climate-sensitive perils.”
Where the industry generally does agree is that a component of loss growth reflects construction trends, whereby developers have invested in recent decades in popular coastal sites and in building on land that’s prone to flooding and wildfire.
As one broker said: “Thirty years ago there was nothing to Miami, now it’s a big city full of expensive real estate.”
Conversations about the extent to which climate change may be changing the risk profiles of extreme weather events tend to fragment quickly into the details of how well the catastrophe models work and how effectively or not specific perils are currently modelled and understood.
The past couple of years has seen the sector generally accept that so-called “secondary perils” - those which tend to hit more frequently but causing small to mid-size losses – are now occurring more often, owing at least partly to climate change.
Secondary perils include floods, severe convective storms and wildfires and are generally not particularly well modelled or understood.
Analysis by Swiss Re’s sigma research operation found that of the $81bn of insured natural catastrophe losses incurred in 2020, $57.4bn, or 71%, were caused by secondary perils. The main components of the secondary peril losses for that year were convective storms and wildfire.
This year, the Bernd flood in Europe is expected to contribute about EUR10.0bn ($11.6bn) to global industry cat losses.
Howden’s report noted that 12 loss events of $10bn or more have hit since 2017, “more than double anything previously seen,” with secondary perils “contributing significantly to the trend.”
Catastrophe risk modeller Karen Clark & Co crunched the data from the first instalment of the sixth report of the Intergovernmental Panel on Climate Change (IPCC), to reveal that flooding and wildfires were among the perils most vulnerable to climate change, with confidence high that they will increase in number and intensity.
By contrast, frequency of tropical cyclones was not expected to rise because of climate change, and uncertainty hangs around the outcomes for severe convective storms and winter storms.
[Chart – KCC. Please reproduce chart from this story IPCC report points to more intense hurricanes in future: KCC (trading-risk.com)]
Work is underway within the industry to agree on appropriate methods of stress testing what could happen to a given natural peril under certain climate conditions, but this is in its early stages.
With 2021 shaping up potentially as another above-the-long-term-average natural catastrophe loss year, the jury still seems undecided on what are the right ways to frame these figures.