Don’t oversimplify climate risk model results: climate panel

When it comes to using catastrophe and climate models, of all types, to assess risk, it is important to understand the potential variations and not oversimplify the results, which can be very nuanced and require sophistication in how they are interpreted and used.

That was one of the key messages that emerged in a panel discussion held at Intelligent Insurer’s  Climate Risk & Sustainability Europe 2023 conference, being held in London today (June 7), called ‘Boost climate resilience for your clients with risk management services and innovative products’.

The session featured Garth Marshall (pictured up), head of climate change and sustainability, Zurich Resilience Solutions; Yingzhen Chuan (pictured botton), regional director, head of climate and ESG, Gallagher Re; Adrian Champion, climate research lead EMEA, Reinsurance Solutions, Aon; and Luis Sousa, catastrophe risk and partnerships lead, Mitiga Solutions.

“We must be careful to not oversimplify what is a very complex message – it needs to be translated,” Champion said. “We will inevitably ask things like what does it mean for my portfolio? And some may translate it and want to increase the premium, perhaps. But that is often too simple. We have to navigate the great uncertainty within the data with specific solutions people might be seeking. We have to bear in mind, this is rarely solved by a single value.”

Sousa explained that Mitiga focuses mainly on secondary perils such as wildfire risk and weather-related risks including drought. “We approach it from a different perspective to the traditional cat modellers,” he said. “We start with climate modelling and apply that to weather and present and future hazards. A challenge now for the wider industry is assessing frequency and severity of events – so we are trying to take a step back and apply what climate conditions mean for frequency and severity.”

Champion at Aon added that catastrophe models are a very useful tool in terms of assessing the current climate and noted that global climate models offer more top-level information. But it can be difficult to apply the latter to specific events, such as hurricanes. “Science is constantly evolving but a lot of the commentary on climate change looks at the top-level hazard – we need to translate hazard into risk.”

Chuan at Gallagher added that the idea of climate resilience should be understood in a way that goes beyond direct damage to assets. She said casualty modelling is interesting in this regard. “It is not a question of using just what you can take off the shelf, you have to dig into the models. Many tools need to be used on top of the models alone.”

Marshall agreed that while insurers look at natural hazards as they are now and have been in the past, it is now a question of applying climate data to that. “That can be very valuable and should be used in addition to observations from site.”

Sousa also described the inherent uncertainty. “But how do we make sense of that uncertainty and in a way that is actionable? What is the use case?” he said. “Also, when you consider how climate models would have predicted the climate in recent years, we see that actually multiple models perform much better than one single model. That is helpful to know but you also need to identify what you are trying to understand. And you need to have a range that accepts there is uncertainty.”

Marshall considered how re/insurers can then use data in the real world. He described working with clients on risk mitigation and climate resilience. “You cannot prevent everything, but some mitigation might be possible once you have the data and know where to look. That is really important.”

Did you get value from this story?  Sign up to our free daily newsletters and get stories like this sent straight to your inbox.