Catastrophe modelling has become a vital tool for insurers when analysing the scope of their risk, and risk modelling technology is constantly evolving to offer a more detailed picture of various risk profiles.
According to Robert Muir-Wood, chief research officer at RMS, catastrophe modelling can be vital in allowing insurers to quantify risks and determine likely costs – however, it also faces the constant hurdle of uncertainty the further ahead it attempts to look.
“Catastrophe modelling is the original insurtech,” Muir-Wood explained. “It is the ability to put together the hazard, exposure and vulnerability and measure and quantify the risk cost, and we can use this modelling to plan how an entire city can manage its risk.”
“One can think about the consequences of climate change as those that are acute, and happen in terms of short and long-term scale events, and those that are chronic and yield a systemic impact,” he continued. “Events like floods and wildfires are acute – they happen fast and result in physical consequences, and chronic consequences are things that take place over a long period of time, such as sea level rises.
“We then have the systemic consequences of acute events, things like climate change disruption to infrastructure, impoverishment and migration.”
Muir-Wood says modelling techniques can be used to map a vast array of potential events, but the results produced depend heavily on the timeframes being explored. Modelling geared towards different sectors might product different results, which makes it difficult to produce a perfectly accurate picture.
“We try and model risks for things like heatwaves, fires, droughts and floods – extreme rainfall, rivers, tropical cyclones, etc.,” he explained. “Climate change is already shifting the severity of these perils, and we need to take into consideration how we price and manage our insurance and reinsurance.
“Another key feature of the future is that it becomes more and more uncertain the more you look into it. If we’re looking at insurance, we’ll look at a time horizon that is principally 1-5 years into the future, which is quite short-term. If you’re building a house, you’ll take at least a 20-30 year horizon, which means you’re likely to get a different answer. If you’re building infrastructure, you should be taking a view that’s at least 50 years into the future. All these answers and levels of uncertainties are going to be different.”
Ultimately, Muir-Wood says RMS’s goal is to provide as good a risk model as they possibly can, and this involves taking this vast patchwork of factors into account and analysing them in the most effective way possible.
“Our simple mission is to deliver models that best represent risk over the relevant time horizons, and that’s really critical to what we try and provide,” he stated.
“We need to include our thoughts on changes in the hazard, and also changes in vulnerability. The consequences may change in terms of increasing litigation; we find in Florida is that the increased involvement of lawyers in claims disputes is having a dramatic effect in pushing up costs of events. We need to be on top of all these issues when modelling the cost of risk.”