4 questions for Hyeji Kang, Head of Reinsurance And Catastrophe Risk Management

The pandemic has pushed insurers/ reinsurers to think more carefully about “black swan” events which we didn’t think were likely to happen. This challenged many of us in the industry to develop new ways to identify, assess, mitigate and manage these risks – something in which reinsurance plays a big role. At AGCS, we worked with colleagues from multiple insurance functions to discuss exposures, how reinsurance would respond to various scenarios and how else we could mitigate the risk. Given the heightened awareness and focus on topics like communicable disease and cyber events, coupled with the hardest reinsurance market since 9/11, the biggest challenge we had was to secure proper levels of coverage and capacity given our own retention appetite on certain covers, as reinsurers were focusing on coverage and wording topics – even more so than pricing – especially for lines exposed to pharmaceutical, cyber and contingency risks. The ways in which we handled reinsurance business had to change dramatically. The industry managed quite well in the end, but the slow pace of renewals was due to the combination of remote working and a hard market.

In general, for both modeling vendors and insurers, the challenge is to keep up with the ever-changing list of major risks insurers face and to learn how to correctly predict them. the pandemic has showed us that the exercises insurers conduct to identify and assess potential risks can sometimes miss the magnitude and frequency of similar global events. the industry for some time has been developing ways to assess cyber risks and understand what kind of damage the next wide-reaching cyber-attack could do to their portfolios. When we invited several vendors to show us how they go about it, we found that the topic is multi-faceted. Depending on what you focus on (pricing, accumulation analysis, etc.), the model outcome can widely vary. For insurers, the issue is more daunting because we must take this information and make a decision on how we shape coverage, how we price it and which loss mitigation controls to consider. these complex man-made accumulation events are particularly difficult to model because there are human factors – unlike modeling a damage from a hurricane – both in the cause and prevention of the events and the claims in the aftermath of an event.

Speaking of human factors, another modeling challenge for the industry is to understand the impacts of climate change. Not only are the impact of the weather and rising sea levels hard to predict, but there’s also the human behavioral change – such as urbanization causing a greater potential for flood damage and growing abstraction of fresh water from rivers and groundwater resulting in increased risks from drought, for example. The industry is only beginning to learn how to encompass all these factors into a perfect model, but because climate change happens gradually the accuracy of a prediction for any given year is more difficult to rely on. I think what insurers can do amid the uncertainty is to actively focus on working with customers on preventative measures and steering our portfolio composition with these numerous not-yet-well-quantified risks in mind. Our risk consulting is well positioned to help in this sense.

Speaking of human factors, another modeling challenge for the industry is to understand the impacts of climate change. Not only are the impact of the weather and rising sea levels hard to predict, but there’s also the human behavioral change – such as urbanization causing a greater potential for flood damage and growing abstraction of fresh water from rivers and groundwater resulting in increased risks from drought, for example. The industry is only beginning to learn how to encompass all these factors into a perfect model, but because climate change happens gradually the accuracy of a prediction for any given year is more difficult to rely on. I think what insurers can do amid the uncertainty is to actively focus on working with customers on preventative measures and steering our portfolio composition with these numerous not-yet-well-quantified risks in mind. Our risk consulting is well positioned to help in this sense.

Catastrophe risk analytics at AGCS is already actively used when we try to price the risks and also monitor any significant accumulation from a weather event. We provide services called “client risk profiles” for our clients with many global locations, using such analytics and insights. This service helps customers know precisely what their locations are susceptible to – such as flood risk for buildings near water, hail risk for flat, large roofed buildings or wind risks for multi-storied buildings – and what insured values are exposed, which then leads to better risk management measures. This goes beyond a simple property insurance implication – an investment firm with a large real estate portfolio would also be keen to know and protect these assets on behalf of their investors. There are also other by-products coming out of the tools we have developed to map the concentration of insureds. Some non-weather events, like terrorist attacks, riots, environmental contaminations, or large explosions like we saw in Beirut or Tianjin – basically anything location specific can be simulated and monitored. These tools are key to AGCS’ own risk management, and provide our clients with meaningful insights as well.