Reinsurers’ new climate risk reality

Climate change is a reality that reinsurers must respond to, especially on the underwriting side of their balance sheets.

Nearly three-quarters (71%) of reinsurers replying to a survey by Standard & Poor’s said they considered climate change in their pricing assumptions – but only 35% said they include a specific component of the price allocated to climate change.

In a sobering conclusion, S&P said that while reinsurers have increased their efforts to incorporate climate change into their risk management processes, it is still “nascent” across the industry.

Aon’s head of catastrophe insight, Steve Bowen, agrees that climate change risk is an evolving and challenging issue for reinsurers. While physical risk remains the most obvious point of focus as the intensity of weather-related disasters becomes more evident, the totality of climate change risk for the reinsurance industry goes beyond the direct impacts on the property and casualty sector, he says.

“The interconnected nature of the global economy and cross-sector workings of many industries has only amplified the types of non-physical impacts which can be incurred far downstream from where an individual event – or set of rapid-succession events – has actually occurred.”

Bowen says reinsurers are looking at making fundamental changes to how their underwriters view the compounded effects of climate risk, as well as the way to properly price or account for how events may be affecting broader portfolio exposures.

Progress – of sorts Olivia Brindle, head of sustainability at speciality insurer and reinsurer Fidelis, thinks that there is more progress on physical risk than transition risk because the former is relatively easier to model and fits better into existing frameworks for both underwriting and risk management.

“Having said that, I have serious doubts as to whether the industry overall is really reflecting even physical risk adequately,” she adds. “If you ask most companies what their loading for climate change is, I’m not sure you will get a robust answer. On transition risk, while this may perhaps be captured in risk management scenarios or stress tests, it is certainly not well incorporated into underwriting decisions in the sense of portfolio steering or individual risk selection or pricing.”

Fidelis group chief risk officer Phil Murfet concurs, saying: “Recent exposure pullbacks and associated pricing momentum appear more driven by a lack of reinsurer profitability, reduced retrocession capacity, fear over inflation and a flight to quality than a tendency to increase the baseline view of risk for climate change.

“Some markets point to facts like the number of hurricanes not increasing statistically significantly over the last century as enough rationale to ignore the projected impacts. Some also reason that given the cat business is [typically] placed annually, there is no need to make wholesale changes to risk views.

“While it is fair that the mean climate doesn’t get materially worse year by year, the odds of the most extreme-weather events increase exponentially,” Murfet adds.

Improving catastrophe modelling is an important area of focus for the industry, with carriers’ in-house teams, brokers and third-party vendors all working hard to produce models that reflect a fast-changing climate. The vendors are all building new data into their reference models to take account of, for example, changing hurricane severity.

Joss Matthewman, senior director of climate change product management and strategy at RMS, says the uncertainty around climate change is huge because there isn’t sufficient historical data available.

“Perhaps counterintuitively, for many extreme-weather perils we can have greater confidence around climate change signals in future than for a historical period we’ve already observed,” says Matthewman. “And the challenge now is how to use that information to better inform our understanding of potential climate change over the past few decades.”

Impact Forecasting, Aon’s in-house catastrophe modelling unit, has multiple academic partnership agreements to directly implement climate change research into its global catastrophe model suite, Bowen says. “By working directly with academic researchers to develop and implement future climate change event sets at various time horizons using different scenarios into the model, this can aid reinsurers in understanding where the risk may evolve next year or 50 years in the future. The work uses a downscaled approach to help better pinpoint where risks may be changing,” he explains.

While computational power and data availability remain a challenge in providing believable results at an asset-level location globally, the models are making progress on delivering an idea of what reinsurers may face in the future, Bowen adds.

Fidelis’s Murfet says model vendors have so far only provided alternative views of risk “conditioned” with climate change and have been reluctant to make changes to their baseline views of risk.

“This leaves reinsurers having to fend for themselves when it comes to determining the right approach,” he explains. “Even with an updated view from vendors, or one generated internally, it is structurally difficult for organisations to impose an enterprise-wide view of risk across multiple classes of business and underwriting centres that are competing for the same capital.

“Early adopters of more prudent views of risk will naturally derive higher prices and internal capital requirements, meaning the only logical trajectory is to reduce catastrophe exposure. Hence it will be tough for those reinsurers to remain competitive until there is enough pressure from regulators and rating agencies, or further years of limited profitability, to move the rest of the market,” Murfet adds.

Model vendors need to support their clients in the context of climate risk, according to Matthewman.

“We all need to better understand what use cases our clients have and what their future needs will be. That applies to regulatory requirements but also to others.

“More advanced clients are already outlining their use cases to us – requesting, for example, a present-day view and understanding the uncertainty around it; what their portfolio risk will look like in 10 or 15 years’ time and a solution for evolving regulatory requirements.”

Matthewman says it is common for reinsurers to use tools that do not align with their current decision-making. “The company has developed a view of climate change, but it doesn’t know how to use it and translate it into loss projections and decision-making around risk appetite and capital setting, for example.”

Reinsurers do face modelling challenges for both physical and transition risk – although, on the physical risk side, adjustments to existing practices should be sufficient, says Olivia Brindle. “Fidelis does just this, incorporating estimated climate change impacts for all relevant lines of business. Clearly these are always just estimates, but we know they at least go in the right direction and we have explicitly considered them.”

On transition risk, the challenge is that a new perspective is needed on portfolio steering and risk selection, with less emphasis on certain traditional sectors and much more on new technologies, some of which are emerging and relatively untested, Brindle suggests.

“I think there is also uncertainty about how fast to move – a fear of making changes that are seen as being too drastic too soon, not knowing how the trajectory towards a greener economy should really look, and as a result deferring taking any decisive action,” she explains.

Climate change requires complex solutions for reinsurers to not only manage the risk but also lead in the space, Bowen says. “However, the industry is just coming out of the starting gate in terms of how to meet the growing demands from such groups as global regulators, boards of directors or activist investors.”