Aon plc (AON) Earnings Call Transcript & Summary

No. I mean I think Pradeep covered it really well but I think one thing he said that I think is absolutely right and it's a very fair criticism of the insurance industry, is how short-term focused we tend to be on risk, especially physical climate risks. And there's a very real disconnect between the insurance annual policy, how insurers typically look at. They're trying to make a profit next year in their rates and in their methods versus on the investment side and on the lending side, which is a much longer view of profitability. I think that from a planning perspective, there's so much opportunity for the industry to use these longer-term views and really get quite specific in how time at change will affect profitability, will affect their growth plans, will affect their portfolio distribution over time and expose some of that information more transparently to the asset owners and to the lending community. On the -- I was going to bring this up in a later question, but it's appropriate now when the Fed did their climate stress scenario for the 6 largest banks last year, just last week or 2 weeks, they released their kind of findings and their summary findings from that. One of the consistent key issues raised by every bank that went through that exercise was the question of insurability -- insurance affordability over the duration of their investment was a key factor in how material climate change it really is to their underwriting portfolio, their loan portfolio. So we have to help to answer that question. We have to help make data more transparent over a longer time horizon in order to make sure that all of the stakeholders that are deciding where to build, how to build, where to invest, that they have the same kind of view of what the risk will be.

Yes. I mean at the end of the day, this is -- it's definitely an affordability issue. And I think all of us even here that are based in the U.S., I mean, all of our premiums are going up, whether or not we're filing a claim. But what's really going to be the growing problem is the availability issue, right? And it goes back to how much pressure do we want to put on the public insurance entities versus the private market, right? Because the prior market is trying to figure it out, it is putting more strain on the states. And we've already seen legislation passed in Florida. We've seen it in Louisiana. The states are trying to get a handle on. Some states are calling it a crisis, right, in some respect. So how they are handling all of this is going to be really important, and it really does require more of that, that private-public partnership to figure out what's going to be the most realistic way forward because it's very close to spiraling out of control. If we have particularly a hurricane season that may be quite challenging to the market. So stay tuned.

Thanks, Steve. Can we talk about those -- the long-term exposures and how can insurers better incorporate climate data into managing their long-term portfolio planning? And Steve, would you start us on that?

Yes. I mean I think this is a really interesting question. Actually, I'm going to be curious to hear what other panelists say because the way that insurers are thinking about it still really the next 365 days, right? I mean that's a traditional business model that we're only looking at the next year. But as more and more financial segments are starting to look longer term, and as you're seeing regulators start to ask questions about how you might be affected based on your current portfolio exposure based on future environmental or climate hazard scenario conditions in the next 20 years, 30 years, 50 years from now. I mean that is really starting to philosophically change how insurers or other financial markets are starting to think about their own portfolio risk, right? So I think this goes back again to the quality of data. I think that a lot of times, folks are just happy to get their hands on any data, but not all data has the same quality, right? I mean it depends where it's coming from, what is the type of data? Is it actually going to answer the questions that you need to solve the internal questions that you're asking yourselves. So I would say that I think we're definitely getting better. But because there are so many of these third-party data providers that are increasingly coming onto the market, and I do want to say that not all third-party data providers are nefarious, that they're not actually providing a good service because there are some really good providers that are out there. But this goes back again to the education piece because a lot of the data that's being used by these companies, that are being used for these longer-term portfolio planning are effectively based off all the freely available data that's coming out of Noah, right? Or the JMA or Copernicus. So having that in-house expertise is going to be really critical in terms of identifying specifically what it is that what you're trying to answer.

Pradeep, could you weigh in? I know you've had some interesting information looking at wildfire risk in California and flooding in New York? And how are you measuring the long-term implications of climate?

Yes, of course. So as mentioned, the longer-term implications are quite substantial, especially for companies that have -- investors that have a rather longer time horizon to look at. Of course, most models that I have seen to date are probabilistic, are decadal in nature. And so you're looking at expected estimates over a much longer time horizon. That can be very important, but it's also important to not overly rely on -- certain overly expect certainty from external models? Any third-party provider that can tell you or that claims to tell you they can accurately predict a hurricane or a wildfire 10 years from now is ambitious, to say the least. And so it's important to understand that similar to many credit and other loss estimates, these are all probabilistic. And so you need to be able to analyze the data with that view in mind. And this is something that we're doing in-house with larger big data techniques and other simulation estimates to try and really drill down into getting the loss estimates, the distributional impacts of these longer-term climate events.

Yes. Maybe to add on that. So for us, as you mentioned, Pradeep, we need to run all the probabilistic models, and these determine our risk appetite as of now. I think for every insurer, amount to determine that risk appetite that everyone has in high-risk areas. So accumulation risk management is key for everyone. And I've had a lot of interactions over the last couple of months, also with lots of banks and asset managers. And they do not have kind of the privilege to price on an annual basis only. They have to take into account a longer period of time. So they need to look into 20 years ahead or 30 years ahead. In P&Cs or in property casualty business in non-life, we do renew our contracts on an annual basis. Yes, we do drive all our models with a climate signal already. So that's what we constantly improve our model landscape and our environments that climate change is already part of that. On a different note, maybe on a related note, just last week, I've had interactions with lots of life and health insurers. And from their perspective, it's already a pretty topical topic as of now. They need to think about how does their longevity and mortality rates will change over decades. This is something that's on top of mind for them. It's not that material and business risk as of now, but they are preparing already. So it has to do with an increase of heat waves. It has to do with the spread of infectious diseases, even mental health issues, which may arise out of climate change is becoming a topic for them? So that's something that is becoming also increasingly relevant.

Yes. I mean just to jump back in. I mean the whole concept of compound risk or connected extremes, I mean, is really becoming a much more important factor in terms of any type of risk assessment, right? I mean -- a few of us have already talked about the downstream implications where I think we tend to focus more on just the physical risk component of all this, but we don't necessarily think about health. We don't think about supply chain. We don't think about all of these other bigger factors that may not necessarily be called a direct loss or direct impact, but they are really leading to, again, philosophical changes in terms of how companies are identifying risk and how they're being affected. But making those connections are going to be increasingly critical as we continue to see a more globalized economy. All of these things are interconnected. I mean, even the geopolitical perspective of all this. I mean you look at the Pentagon. I mean, they reference climate risk all the time in terms of how they're planning, but you talk to layman on the street, and I don't think most people are connecting with Pentagon with climate change. But I think it just speaks. That's just one example that speaks to how far down the road the implications of all of this actually means. So how we communicate that, how we recognize where the connected links are and try to help solve that and help develop some type of strategy from an advisory standpoint, I think it's all going to be pretty critical in terms of how you're actually looking at your portfolio and seeing where the risk actually exists.

Steve, can I draw you out a little bit on that? How -- could you just connect the dots for me? How is political or civil unrest related to -- or potentially related to climate change?

Sure. Well, I mean a lot of it can be driven by, say, food and security, right? You get certain areas of the world where they're seeing extreme droughts, and there is this massive need for food. So there's entire segments of people. They're referring to as climate refugees that move into other parts of the world, especially in, say, the Middle East, where you get various religious sects that are now integrating because people are just trying to seek food for survival, and that can lead to more unrest. So that's just one example in terms of how this is all tied and how the geopolitical component of all this is a very real thing and consideration that governments are increasingly paying attention to.

Liz, I see you nodding. Did you want to weigh in on this, too?

Yes. I mean, obviously, the impact of climate goes beyond the acute perils that we're sort of familiar with in the insurance industry and climate-related migration is something that can lead to quite a lot of civil unrest, political instability, changing dynamics, just to broader economic issues and country security issues that we're already seeing the impacts of. I think that one question that came through the chat was around what are some of the things that you would think about today if you were starting an insurance company from scratch. And to my mind, I think something that the industry doesn't really think about too often is the impact of these kind of chronic perils on their business, on their customers and on the environment in which they operate. If you just think about heat stress as an example. Heat-related events are going to continue to increase, are already increasing and have really diverse impacts on people, on properties and on our businesses. The heat waves from last year in the Southwest U.S., Phoenix had 31 days above 110 degrees Fahrenheit. In prior climate models that was like a 1 in 250 event. Taking into account climate change, that type of event is expected to occur once every -- 1 out of every 15 years or 1 in 15 events. That increase in frequency is something that isn't going to slow down. Every climate signal shows that heat is expected to increase. And it's the most easily understood payroll change from a data and analytics perspective. It's hard to say how climate change will affect hurricane behavior, but we know how it's going to affect heat and heat waves. He will affect people's cardiovascular health, their mental health, their ability to work for every day above 40 degrees Fahrenheit or 40 degrees Celsius, we lose like 72% of the workers' productivity. Some estimates say that the U.S. is already experiencing about $100 billion impact to GDP because of heat today and that can increase up to $500 billion by 2030. Heat is going to affect health. It's going to affect life insurance policies. It's going to affect our company's ability to stay open and operate. It's going to affect your company's employee well-being. And it will also affect P&C. When heat waves occur in parts of the country where the infrastructure is not able to stay open to stay running to keep the cooling HX systems working, those are going to become complex insured losses. Company that we partner with, who models these scenarios looked at on heat wave event, that could cause above $120 billion of loss to the insurance industry through workers' comp claims, general liability claims and workers CFD claims. So this is something that's going to be far more tangible and companies are going to already feeling the pinch from heat and it's going to increase in the near term. And that's something I would pay attention to right now as an emerging risk that we don't quite have the tools to think about at the moment.

Thanks for that, Liz. It's very, very interesting. Tobias, I'm going to come back to you? And how do you see insurers' use of climate data evolving going forward?

Yes. I mean we touched upon that already a bit. So we, at Munich Re are constantly monitoring, of course, our model landscape and every new model that is going to be revised or compiled from scratch is considered with a climate impact? Or it is monitored to what extent climate change plays a role already today. So climate change is becoming an increasing topic. But the big challenge is we and also all of our peers a bit of struggling in quantifying the real impact. So what is the amount to what we can attribute climate change to play a role with respect to frequency and intensity. The devil is a lot in the detail. So we have the flexibility, as mentioned in the annual adjustments of the models, but it is taking into account absolutely. We just need to map out the difference between what has happened in the past, what our loss experience says us and what do today's risk says and what future risk is. So the view into the back-view mirror is often not representative anymore for what's lying ahead. So we need to understand the different mechanisms and the drivers of these changes that keep pricing accurately and adequately. So higher and better resolution of the data is highly appreciated. It helps us a lot at the end. It's all about also this compound risk, as Steve already alluded to. That's an increasing issue. So multiple events occurring simultaneously that gives lots of more challenges to our industry. Yes. So it is a topic that is increasing and everyone needs to comply with it.

Yes. I would say that in terms of future climate data, if you're doing [ entitle ] modeling, the one thing that we are very much communicating to our clients is don't fall in love with a single model, that you need to have a multi-model approach, you should be looking at various and salable models. It's really important to make sure that you're appropriately balancing because any type of future climate model, GCMs that are being used, there are varying levels of sensitivity, different parameters that are being used to adjust what and how that's going to actually simulate, what future environmental conditions are going to look like. And so falling in love with one mean model is not going to probably be the best use of your skill sets, I would say. So implementing a multi-model view is going to be really helpful in terms of getting a more accurate view because, again, there is so much uncertainty. And the farther in the future that you go, the more uncertainty that exists and I forget someone on the panel I referenced this earlier, but if any data provider or any type of adviser is coming to you and talking about future climate risk in absolute terms, I mean, for me, I immediately just shut down and basically walk out the door because I'm not necessarily going to take you seriously because nothing in the weather and climate realm deals in absolute. So there's always some inherent uncertainty that exists. And if you're trying to communicate what future risk is going to look like at a specific asset location on May 29, 2074, they're not talking to you in good faith. They're not giving you good estimates. So you have to be talking about the uncertainty, you have to be using a wide range of different types of data and different types of future modeling solutions to make sure that you have a fledged view of risk.

Pradeep, I saw you nodding along to that. Did you want to jump in?

Yes, of course. Certainly, one of the things that Tobias mentioned that a way in which we see climate data evolving is looking at correlated and coincident hazards. So the idea that a drought will affect your wildfire risk is intuitively pretty obvious, but not one that we have yet to see directly integrated into many climate models. Similarly, hurricane impacting coastal flooding, resulting in coastal flooding or pluvial flooding. Again, intuitively, they will go together. But these are areas that significantly more model development and analytical research are needed before we come to a place where we can confidently predict even a distributional impact from multiple events. And so I think that's an evolving frontier of use of climate data.

Liz, would you want to weigh in?

I mean I wouldn't want to continue to repeat what everyone says, but I think that the panelists are right about the uncertainty and the challenge around making sure you've got a multi-model view of the risk and understanding to how the science is continuously evolving. So one thing that we talk to our clients about is how to develop really a framework for assessing the models that you're looking at. So before you even get to the point where you're running a model and you're looking at the output, actually develop a framework for how you're going to assess those models. Understand what assumptions that they're making without judgment that one is better than the other, but just understanding how their underlying assumptions impact the output you're getting. And then continuously evaluate that framework over time as the science evolves. We've talked about SCS events in the U.S. and the impact of climate on those. I'd say that's an area of sort of less scientific consensus around exactly how climate will affect the frequency of tornadoes, the strength of tornadoes, where they might occur. So you might take a more nuanced approach to making any adjustments to your models for that type of peril versus flood or heat, where there's a stronger linkage and there's stronger scientific consensus. So that framework is important to establish initially. And then you continuously evaluate against it. It also can help you when you -- when it comes to disclosures, and I know we've got a disclosure question coming up, but the governance on how you're evaluating the models, how you're using them, what is your framework for reviewing those assumptions and adapting as research evolves is a critical part to protecting your organization as you're making disclosures and you're doing those in good faith.

Thanks, Liz. And let's get to that question. And we touched on it earlier, but what are the impacts of the proposed SEC regulations on product innovation, modeling and data?

Yes. I think the SEC rules that came out earlier this year were far less meaty I think, than what many on the climate activist side were hoping to see. They're certainly less restrictive than the [indiscernible] regulations in Europe. I mean as an individual state, the requirements that California is putting on businesses for disclosure are stricter than the SEC regulation. The regulation is also seeing a ton of litigation. So we don't really know where it will fall, what will ultimately be in those rules, and of course, the future is very uncertain given the election cycle this year. So it's hard to say exactly what the impact will be, but what I think is not going to change is that disclosures in some form will start to come for U.S. publicly traded companies. They will be expected to create that governance framework in-house. It's going to require organizations skill to understand how to use climate models and also how to understand the impact their business has on carbon emissions and the impact that their business can have on increasing or reducing those carbon emissions over time. Data quality is going to be critically important as these disclosures continue. And one thing that we see as a real concern from the clients that we talk to, especially at the board level, is what risk they're exposing themselves to as they make these disclosures and as they make commitments to various net 0 ambitions and their ability to actually achieve or deliver on those commitments. In terms of product innovation, they'll be -- I think the insurance industry has an opportunity to really lean in and support organizations who are making those net 0 commitments, organizations that are making bets on carbon capture, on nature-based solutions, on things that help to actually offset and reduce the amount of carbon in our atmosphere. Those organizations are looking for protection to do those investments and to make those bets. And so from just like a bit of an opportunity for the industry and a pitch for the industry is to kind of find what those solutions are so that we can enable that transition to occur. Outside of product, data physical data being more precise. Data about our built environment is really lacking outside of the insurance industry. We're really good at knowing if a house has a hurricane reinforced window shutters or a new roof with the right attachments in Florida. We're pretty good at understanding that kind of information and detail about the homes and businesses we insure. Banks and investors, they don't know that information. They don't have that kind of detail because it's not relevant to the credit risk that they've been concerned about to date. So more transparency around the built environment is, I think, also an area where we'll see a lot of advancement and increased transparency.

Thank you, Liz. And we're nearing the end of our time together. So I'd like to go around the virtual panel one more time and ask our excellent speakers to give us what they would like the audience to take away from our discussion today. So if we could please start with Tobias and then go to Steve and then Liz and then Pradeep.

Yes, happy to. And I think the last point Liz was also an important one. You touched on the net 0 race. We do see here some risk already there of greenwashing also in the industry. We need to be very cautious on that. So what is really a material risk to be addressed here on the NatCat and climate risk data topic that we discussed most of the time today, I think our stance is pretty clear. We have an affordability topic that is appreciated, that's highly reflected and accepted in the market. We need to analyze climate and NatCat data thoroughly. We have here at Munich Re lots of expertise dealing with these kinds of changes. We have a long history and heaps of historic data as well. And we have an appetite to grow in that segment. So we will not step out of the market, but it's paramount to run our business model to have risk-adequate pricing. And climate change presents a changing risk. So we need to be on top of that and understand it very well with all the uncertainties that we discussed extensively.

Thank you, Tobias, and Steve?

Yes. I mean this is a pretty familiar frame for me. It really comes down to the communication of risk, right? I think that all of this really translates in terms of how are we as an industry and insurance industry. And I think we are very uniquely positioned. In fact, we sort of sit in the middle, right? I mean, we can work with academics. We can work with governmental entities. We can work with of various financial segments, various financial sectors to really drive innovation, drive conversations in terms of how we actually are turning around and communicating the view of risk, not just at a global level, not just at a regional level, but actually down to an individual asset, where people actually live. And I think that within an industry in terms of insurance, we sit on, frankly, almost more data than anybody else. And as we continue to invest, we continue to prioritize ways of updating our hazard maps or identify just other types of modeling output to be able to showcase to all the way down to an individual policyholder what the risk looks like, where somebody actually lives. I think that can really be a pretty significant turning point in terms of people understanding what type of risk that they face. What are their houses? And if we're able to get more investment from the Federal government or the states or even private sector investment. Deals with some relationships with the World Bank continuing to take off or the United Nations, where this funding comes into place where we can start to help people that may not have the means to protect themselves. But if they understand what the risk is, maybe that will help push forward more of this financial protection to ensure that we are better prepared for this increasingly risky world.

Thank you, Steve. Liz?

Yes, I mean, great panel, and thank you for the discussion today. If I were going to say something, maybe additive, I think one thing is I see a lack of innovation really in the insurance sector to lean in and solve for some of the climate-related challenges more broadly than physical climate risk. I think that we should take comfort in the fact that the models we've relied on to date, while there are challenges and there are gaps and there are needs to improve, that they've led to a fairly sustainable, resilient industry over the long period of time. But we're not at the table around the energy transition. We're not really in the room when we're talking about the type of investment needed to achieve net zero. How are we as an industry protecting nature? How can we close the protection gap? How can we bring insurance coverage to parts of the world where it doesn't exist right now. How can we support regenerative agriculture, carbon capture and storage technologies, renewable technologies. Those are things that are so critical in order to offset the impact that climate change will have on our own results. It's mutually beneficial for our industry to help to solve that because if we can reduce the carbon in the atmosphere, then we can prevent the increase in climate-related hazards in the future, which, of course, is something that we want to see. So I think that call to action for me is really important for people to take away.

Thank you, Liz. Pradeep.

Yes, thanks. I think that we're going to -- on the data side of things, we're going to get better at both analyzing climate data and predictions, but also in terms of the education piece, just speaking about these risks going from this is the expected loss to, this is the probability of loss or the climate value at risk or the distributional impacts. And I think that's a market education piece that we'll really see take off in the next few years. I do also anticipate more government involvement and interaction, both in terms of regulations, but also in terms of government insurance pool starting to build up with the attenuated, moral hazard and other issues. Yes, I certainly would expect a little bit more regulatory enrollment as well. And so definitely an exciting place to be in the next few years.

Thank you, Pradeep. And I'd like to thank our panel, Tobias, Steve, Liz and Pradeep for sharing your insights, and thanks to our audience for tuning in. We'll have a replay of the webinar available in just a few minutes. For Insurance Insider, I'm Meg Green.