Antony Ireland, Founder, Better Insurance Network (AI): Shall we perhaps just set the scene, get a bit of context, and can you just explain what we’re talking about when we’re talking about view of risk and then bring into that why it’s so important that we incorporate climate change into that?

Liz Henderson (LH): Absolutely, and I think that’s such an important place to start because one of the things that we’ve seen at Aon is a wider set of clients, not just in insurance, but outside of the insurance space as well, who are looking for support and help in understanding their climate risk. And the concept of view of risk is so fundamental to everything that we do in insurance to really price and underwrite and create profitable portfolios.

What a View of Risk really means is that an organization who is making decisions based on risk data, whether it’s from a Cat model or an underwriting model or other data sources, they have to ensure that the data that they’re using is properly representative of the actual risk that they’re taking on and represents their own understanding of their portfolio, understanding of their claims practices and pricing practices and the type of exposures that they’re bringing on.

And so a View of Risk and having a framework that allows a company to develop a View of Risk gives you the tools to understand all of the different factors that might affect loss outcomes, and apply that knowledge and learning to all of the decisions in your organization that use that data. So for example, you might believe that one model is a better View of Risk for you for hurricane, and you might also decide that I need to layer in some assumptions around non-model sources of loss, like loss adjustment expenses or the impact of climate change or the impact of social inflation.

You take all of that knowledge and learning and then you have to build out the workflows to embed that knowledge into your underlying rate filings. How do you make sure that your filed rates are representing that View of Risk? You build it into your underwriting guidelines so that the type of risk you’re bringing in matches up in areas where you know your view of risk is accounted for and priced for adequately. You put it into your risk transfer decisions so that you can buy the amount of reinsurance and support that you need relative to that same benchmark. You use it to evaluate the cost and the the risk transfer and the value of that risk transfer.

So a View of Risk is really a framework and it’s a way of doing business that makes sure that you are applying the best understanding that you have around climate risk to every aspect of your business. So if you don’t have a View of Risk framework and you don’t have a method to apply your View of Risk to all aspects of your business, then you’re going to struggle and you’re going to find surprises and you’re going to experience more volatility than your peers who have a framework that allow you to pull the different levers that you need to account for a changing risk dynamic.

And it’s not just climate change. You know, if you look at the inflationary environment over the last few years and the impact that inflation has had on losses and on risk appetite, that is part of View of Risk. If you have that framework, you can adjust and apply and and quickly respond to those types of changes and make sure that you come out experiencing as minimal volatility as you can.

AI: What are the first steps to establish, first of all, your baseline today and then bring in those future scenarios in an effective way?

LH: Before you even start to respond to, you know, what does my risk look like 30, 40 years from now? We really do have to take time to assess the current models that we’re using and make sure that they are actually representing the current risk. And I think that, rightly so, the catastrophe models that we’ve long relied on have come under a significant amount of scrutiny when we’re thinking about baseline risk. Because the nature of the research timelines that the vendors use, the nature of how they’re built and developed, they’re always going to be catching up to the latest science and the latest understanding of how risk might change.

However, I do think they also get a bad rap because first, a peril as complex as hurricane for example, there isn’t much scientific understanding of how climate change is affecting hurricane risk in the next near term three to five years. And so the Cat models are probably doing a very good job of assessing that.

But then when you think about a risk like wildfire that is changing much more quickly and is changing in predictable ways, when you think about the impact of climate change on wildfire, that is an area where we think that the models can do a better job. And so the goalposts are always shifting in a number of ways. One is the understanding that scientific community has and can give us around how climate change is actually affecting the frequency and severity of extreme disasters. That area of research is active, it is fast evolving and it will fundamentally change how we apply the scientific learning to our View of Risk.

“What was good enough last⁣ year is not going to be good enough five years from now.”

So you know, five years ago we might’ve said we think climate change will have a nominal effect on Atlantic hurricane activity in terms of overall base and frequency. And now we might say, well, yes, we think the frequency might go up a little bit, but it’s going to go up more in the category three, four, and five storms.

But there’s still quite a lot of research needed to understand, well how does that actually affect landfall hurricanes because basin averages are useful, but they don’t really matter. What matters the most is what events make landfall. And so that complex research is going on in universities right now and we have to be able to adapt to that changing science. What was good enough last year is not going to be good enough five years from now.

The other way that goalposts are moving, regulators are also trying to really grapple with the impacts of climate change. And so globally, there’s a real fragmented regulatory landscape that’s developing that is asking organizations to disclose climate related risk risks against different time horizons, different scenarios, climate scenarios, different perils and different really fundamental methodologies. And so organizations need to have a way to respond to these different requirements.

And then I think just one last thing that people need to watch out for is we focus quite a lot on extreme events – acute perils – that’s an area that really matters to insurers. There’s a lot of capital in Florida hurricane, there’s a lot of capital in earthquake. We need to make sure we get those perils right. But at the same time, we know that more secondary perils, like severe storm and wildfire and floods are changing how they behave, that losses are changing in ways that we don’t quite fully understand and we don’t quite fully understand how climate is affecting those changes.

And there’s also the increase in more chronic perils like heat and heat stress that has an impact on insurers in ways that are completely outside their normal underwriting practice. Heat stress, heat waves, those are more directly tied to climate change than other perils. The science is much stronger. I mean, heat is the impact of climate change, right? More carbon in the atmosphere leads to higher global mean temperature and all the other effects on weather and catastrophes are knock-on effects. We are only starting to really know how heat is going to transform our cities, transform our communities, transform our businesses.

And we’ve started to see the effects of it. I mean, 2023 broke just about every record of temperature and extreme temperature globally. And this is probably going to be the first of many record-breaking years in terms of heat. So an insurance company has to start now to understand how is heat going to affect my business?

AI: Do you have any tips for insurance companies in terms of what good looks like when it comes to the quality of data, where they find the data and making sure they have the right climate risk data for their particular needs?

LH: It’s always been a challenge for insurers to collect really good exposure information. And it’s going to continue to be a challenge not only because of climate change on physical risks, but because of all of the different desires or requirements or ambitions that insurers might have around reaching net zero.

So we’re going to need to collect more information about our insureds related to things that we might not today understand how it’ll affect loss. But we want to start understanding and want to start studying. So can you collect data about emissions from an insured that you’re writing? Can you collect information about their transition plan? Can you collect information on more sustainability-related indices?

And maybe today that’s not going to have an effect on your underwriting or pricing, but in the future it probably will. And so start today to collect information and data that you might not normally be thinking about collecting.

“An organization that can operationalize climate insight is going to be a winner through the transition”

The other thing I would say is on the actual climate data side, you know, there is a wide variety, as you mentioned, of new data providers that are coming out purporting to be able to tell you what your risk is going to be in 30 or 40 years. And I think in insurance, we know to look at that type of claim with skepticism because there’s so much uncertainty even in understanding your baseline risk. There’s uncertainty being able to project that 40 years in the future. There’s an even wider amount of uncertainty.

And so our view at Aon has been to look at a variety of vendors in what I call this climate service provider space. These are these are modeling firms or data firms that are not using historical information. They’re not traditional Cat modeling vendors. So my tips would be to understand first and foremost what is just the fundamental difference in how climate service providers are building their data sets and building their output.

Really make sure you understand that, get into definitions, which sounds boring, but you know, one provider will say they can give you their AAL of a peril at a certain location. But what does that AAL actually mean? Is it the actual mean loss across the distribution that includes the full tail or is it just taking a mean loss of today? And, you know, scaling it up for impacts of climate change, which would have some flaws in it. Is it an AAL that really is just looking at the hundred year or greater events? Because a lot of firms are just looking at that tail and not actually looking at a full distribution. And all of these metrics might be useful, but if you don’t understand what’s actually going into an AAL that they’re creating, then you’re going to start comparing things one to another in a way that might not make sense.

And then I think I would say, you know, really being able to understand what is the state of the science right now that would allow you to make a credible adjustment to your View of Risk. Because at the end of the day, you’re taking all this information and you’re going to do something with it. You want to make a decision that’s different. And one of those decisions might be to make an adjustment to your View of Risk to start embedding this information into your pricing and underwriting and portfolio optimization decisions.

If you do that without really looking holistically at the state of the science and understanding how confident are we that this is accurate today? Because everyone can use information and come up with a number, but if that number is low confidence or low scientific consensus, you probably don’t want to do anything with it. And so we’ve created a framework to give to clients that help them dissect the state of the science for a particular peril in a particular region. How confident are we that we can draw conclusions based on the academic research right now? And then how confident are we that we can actually use those conclusions? And not all perils are created equal and not all regions are created equal.

So looking at that framework and using that as a way to apply the data is also I would say a critical part of beginning to embed climate data into your decision making.

AI: What are the options for insurers in terms of bringing that data into the decision making process? And what are you seeing in the market at the moment?

LH: Everyone would say they’re taking into account climate change, but that means a lot of different things. First, my recommendation would be to use that framework of the state of the science as a starting point so that an organization can say anything that has high confidence and high consensus around the science that allows us to draw conclusions, then we should spend time on that and really understand two things: one, how does that affect my View of Risk for current business that I’m writing?

Let’s say we get the price right for the next three years. What does the state of the science and the new data that’s available tell us about that and allow us to make adjustments. That’s critically important because that piece is what drives profitability for organizations in the near term.

You also will have to take a longer term view, and this is something that I think is a bit more uncomfortable for insurance companies. We’re not used to looking at a longer term View of Risk changes. We certainly don’t want to start pricing for 30-year risk in today’s policies. But we’re going to need to recognize that the transition of our global economy away from fossil fuels to renewables is going to create a lot of social dynamics. Government changes, regulatory changes, consumer behavior changes, risk, you know, just baseline hazard is going to change if you wait for things to start to evolve.

You’re going to be trying to upskill and trying to understand the complexities of climate change far too late. And the other thing about operationalizing is quickly being able to adapt and bring in new data where we’re seeing changes very quickly. And so building that flexibility into your organization is key and staying focused on those emerging risks as they come out is also going to be really important. And so an organization that can operationalize that insight is going to be a winner through the transition.