The Emerging Risks Interview: Mohsen Rahnama, Chief Risk Modelling Officer, RMS

As we report this week, the recent devastating floods in Pakistan continue to make the headlines, with as many as eight million people in flood-hit areas needing medical assistance. And the financial cost is likely to be huge: the UN estimates that economic losses could reach $30 billion.

Should the market prepare itself for losses on this level- is the new normal? Emerging Risks caught up with Mohsen Rahnama, Chief Risk Modelling Officer, RMS, to address this issue:

“If I take a look at the last five years, a significant number of climate-related events have happened across the world: $1.1 trillion of economic losses over this period, and the insured loss somewhere around $500 billion,” he says.

“If you step back and look what happened in 2021, in Germany and Belgium, there were close to $11-15 billion in insured losses. Then Hurricane Ida, which was veering northeast and the contribution of the flood was something like $10 billion. And then you have a China flood model which is contributing a significant amount in terms of structural damage. So last year, you have flood losses of around $80-85 billion. That means it is the most active peril model compared to anything else.”

“What’s driving it is that I can see more frequent types of flood, I can see more severe floods, and the return period for flood is getting shorter and shorter. The return period for flood, which used to be 100 years, is now more like 20 or 30 years. And what is contributing to this is really climate change. “

Managing flood risk

“Do we understand how to manage it? Most of flood cover, especially in the residential line of business, is really dominated by government support, which cannot be sustainable for the long-term. What is needed is a partnership between the private sector and government to try and solve the issue.”

Rahnama adds that each company has their own plan and strategy to deal with flood risk and the allocation of capacity to deal with the flood risk they write, but the most important aspect is really understanding the science behind it, and what is driving it:

“If I look at the bell curve, I can see that severity and frequency are increasing. But I strongly believe that the insurance market is very resilient, and with proper planning and the right strategy and incentives they can deal with this. Sometimes they have to develop new products, which can really help them to price the risk properly, and manage the risk properly.”

He says that this is a very real issue for the market in recent years, given that Hurricane Harvey created an enormous amount of flooding in 2017, with some regions that were safe on the FEMA map actually experiencing losses.

Proprietary modelling 

Has the market become over-reliant on commercial modelling vendors and neglected proprietary modelling?

“Let’s step and look back! On my team, we have about 250 scientists and more than 70% of them are PhD or advanced degree-holders… the individual problem is going to be very different from the market problem. Of course they can use the models as a starting point, and can complement it with research data… it is this data that will become the key element in understanding the risk.” 

Models to one side, he stresses that we really need to look at the bigger picture: “Climate change and sea-level rise are key aspects. The force of a hurricane can really bring a massive amount of water within the coastal region, and if you don’t have a flood defence there of course that is going to be very difficult. Mitigation and flood defence is crucial.”

For the insurance market,” he adds, “avoiding the risk is not the solution: understanding the risk is the solution. And then pricing it appropriately.”

Looking to the future 

So what areas of risk modelling should cat modellers concentrate on next?

“I’ve been in the market 28 years. In the beginning a relatively simple model gave us a solution, then we developed a more detailed model. When it comes to flood modelling, you need to have a better model to differentiate the risk: I call it the higher resolution model. And if we look at the European flood model, it is well known that some companies such as Flood Re are using the RMS flood model as a reference. “

“What are we going to do in the future? We are building the climate condition model which looks at public information from the scientific community and the latest global climate modelling. We use that information and bring it to catastrophe modelling to identify where the problem of climate change can aggregate to the current solution we have… it is about using the models and the data we have to adjust your activity and predict the future.”

“Looking at flood by itself is not sufficient. You also have the inflation issue, and the vulnerability issue, and the combination of them is the overall exposure.”