Baden-Baden Q&A: Aon's Alexander Alabaster

Can you tell me more about the Catastrophe Research Team at Aon?

Catastrophe risk is the largest driver of volatility on most re/insurers’ balance sheets.

In response, Aon has a dedicated team of 36 research analysts to advise re/insurers on how they can create a custom View of Risk to manage their portfolio, navigate volatility, transfer risk, and optimize capital. It is a global team with deep specific expertise across all perils and geographies embedded throughout our major office hubs. The structure encourages alignment on a global basis, while still allowing the team to adopt an Aon United ethos and collaborate with other business units and specialties worldwide. Ultimately, across the whole of Aon, we aim to help shape better decisions for our clients, and the output of Catastrophe Research operating in this way helps us to achieve this goal.

Why is developing a bespoke view of risk (VoR) important to clients?

After evaluating the vendor models and explaining to clients what is driving the differences in the models, our focus is comparing the models to the latest science and recent claims data. We are then able to give our opinion on the strengths and weakness of the various approaches. With this information our clients are then able to adjust the model components to create a bespoke View of Risk that best represents their portfolio. Increasingly, we are seeing that clients are viewing the vendor models as a framework upon which they can hold their bespoke View of Risk.

With this bespoke VoR our clients can then fully evaluate their catastrophe reinsurance strategies and shape their underwriting based on the composition of their portfolios. It also allows them to gain more confidence on risk tolerance thresholds, data assumptions and loss drivers, and report to management with a risk view which is both explainable and defensible. Clients with a strong VoR can also more easily identify growth opportunities and predict pricing impacts and reduce the risk of poor results driven by modelling changes.

Externally, a robust VoR enables clients to articulate the unique nature of their portfolios to their reinsurance partners, encouraging differentiation in the marketplace and helping to navigate volatility and build resilience.

Why has the need to develop a VoR increased in the last few years?

There are a variety of factors that are driving the increased need for a bespoke VOR. The main ones include:

• Multiple recent losses where there has been significant model miss due to non-modelled sources of loss or from the occurrence of events more severe than previously thought possible, such as the Turkey earthquakes.

• Increase in the contribution of historically termed ‘secondary perils’, such as the recent severe convective storm events in Europe and US.

• The proliferation of available vendor models across regions and perils from the major vendors, and from smaller startups increasing the availability of opposing views.

• Regulators have also had a role in driving carriers to develop a bespoke VoR as there is an increased expectation that re/insurers own their View of Risk and be in a position to defend it.

In what areas do you see a need to develop better insights and data?

There are two main areas that we feel add the most value to our clients: Bringing them the latest learnings from our academic partners, and utilising detailed claims data.

In terms of incorporating the latest learnings from our academic partners, vendor models outside of North America are typically only updated every 10 years or so due to the multi-year development time needed. This means that we need to take steps to avoid using Views of Risks that might be more than a decade old in some regions. Through the Aon Research Forum we are able to leverage a scientific network that provides us with the latest views in hazard for various perils.

When it comes to utilising detailed claims data, vendor models typically only have access minimal location level claims data and predominantly utilise industry level aggregated data. Utilising our clients detailed claims to build out bespoke views of vulnerability for them can significantly increase the confidence in the model results.

How relevant has climate change become in your role, and in the output of your work?

Over the past few years we have seen a significant increase in the need to utilise catastrophe models to quantify the potential impacts of climate change. Our links with academia and our deep understanding of the models has helped us to evaluate vendor models that have started to offer forward looking Views of Risk and develop our own in-house solutions. Aside from the major perils that we are exposed to today, there is also a need to develop solutions for peril regions that are not significant in today’s climate but may be material in the future. Lastly, for some perils, there is evidence that the climate has already started to change. For these perils, it is critical that the models account for this trend in hazard when estimating the losses for the current period.

At Aon, we are undertaking research to understand better how the frequency and severity will change, asking specific questions such as:

• Will cyclones become wetter, penetrate deeper inland and occur closer to the poles?

• Could seasonal and/or serial clustering of European windstorms change?

• What will be the impact to river flows caused by changes to seasonal ice melt?

• Is it inevitable that multi-peril interactions such as wildfire and drought will increase?

It is imperative that the industry invests in the development of better insights to manage the change in hazard, and additionally, not ignore how changes in exposure and mitigation will play out and interact to change our View of Risk.

Through this, our aim is to reduce the uncertainty that has been introduced through climate change, to enable capital to be confidently deployed by the industry.