Modeling disaster risk in 2021: a delicate dance

Complex man-made accumulation events are particularly difficult to model because there are human factors – unlike hurricane damage modeling – in both the cause and prevention of events and claims to be made. following an event. (treety / Shutterstock.com)

One of the most important elements of risk assessment and management is disaster risk management, especially as climate change threatens to tighten its grip. For insurers, reinsurance is an important way to help balance the coverage equation.

Here are some pressing reinsurance issues in 2021.

Reinsurance and COVID-19

The pandemic has caused insurers and reinsurers to think more carefully about “black swan” events that we didn’t think were likely to happen. This has challenged many of us in the industry to develop new ways to identify, assess, mitigate and manage these risks, a process in which reinsurance plays an important role.

My colleagues and I have worked with other professionals in several insurance functions to discuss potential exposures, how reinsurance might respond and how we might mitigate risk. With the increased awareness and focus on topics such as communicable diseases and cyber events associated with the most difficult reinsurance market since September 11, the biggest challenge we faced was to ensure appropriate levels of coverage and capacity given our own appetite for retention on certain coverage. Meanwhile, reinsurers have focused on coverage and wording matters – even more than pricing – especially for lines exposed to pharmaceutical, cyber and contingent risks.

The way we run reinsurance business has had to change dramatically. The industry did pretty well in the end, but the slow pace of renewals was due to the combination of remote work and a tough market.

Challenges of disaster modeling

The pandemic has shown us that the exercises that insurers conduct to identify and assess potential risks can sometimes miss the scale and frequency of similar global events. The industry has been developing ways to assess cyber risks and understand the kind of damage the next large-scale cyber attack could cause to their wallets for some time. When we invited several suppliers to show us how they do it, we found that the subject was multiple. Depending on what you are focusing on (pricing, accumulation analysis, etc.), the model output can vary widely.

For insurers, the problem is more daunting as we have to take this information and make a decision on how we shape coverage, how we price it, and what loss mitigation controls to consider. These complex man-made accumulation events are particularly difficult to model because there are human factors – unlike hurricane damage modeling – in both the cause and prevention of events and claims as a result. of an event.

Another modeling challenge for the industry is to understand the impacts of climate change. Not only are the weather and sea level rise difficult to predict, but there is also the change in human behavior, for example, how urbanization causes greater potential for flood damage, and how the increasing withdrawal of freshwater from rivers and groundwater leads to increased risk of drought. The industry is only beginning to learn how to integrate all of these factors into a perfect model.

Climate change occurs gradually, so it is more difficult to trust the accuracy of a forecast for a given year. I think what insurers can do in the face of uncertainty is to actively focus on working with clients on preventative measures and on steering the makeup of our portfolio with these many risks in mind. still well quantified. Hyeji Kang (Hyeji.Kang@allianz.com), Global Head of Reinsurance and Catastrophe Risk Management at Allianz Global Corporate & Specialty.

Hyeji Kang ([email protected]) the Global Head of Reinsurance and Catastrophe Risk Management at Allianz Global Corporate & Specialty. The original version of this piece first published in the Spring / Summer 2021 issue of the GATS Global Risk Dialogue and is republished here with permission.

See also:


Source link

About Gail Mena

Gail Mena

Check Also

Facebook AI Introduces “Anticipative Video Transformer” (AVT): An End-to-End Attention-Based Model for Anticipating Actions in Videos

Source: https://facebookresearch.github.io/AVT/ Every day people make countless decisions based on their understanding of their surroundings …

Leave a Reply

Your email address will not be published. Required fields are marked *