WEBVTT
00:00:07.200 --> 00:00:12.720
Welcome to the latest edition of the Maximum Exposure Podcast, brought to you by Emerging Risks.
00:00:13.119 --> 00:00:25.440
Ben Rose, president and co-founder of Supersed, discusses with John Guy why re-insurers are having to reassess their approach to risk as exposures become ever more interconnected.
00:00:27.839 --> 00:00:37.359
I'm delighted to welcome Ben Rose, president and co-founder of Supersed, with me on today's edition of the Maximum Exposure podcast brought to you by Emerging Risk.
00:00:37.920 --> 00:00:40.880
Ben, as I said, welcome to the podcast.
00:00:41.119 --> 00:00:43.920
It's been a really interesting year for 2025.
00:00:44.159 --> 00:00:57.200
We saw a on the natural peril side, certainly, we saw the rise of uh secondary perils, but fortunately, I think for the insurance market, we had a very subpart North Atlantic hurricane season.
00:00:57.359 --> 00:01:00.799
However, I suppose the industry cannot really rest on its laurels.
00:01:00.880 --> 00:01:17.760
I mean, from your point of view, on a portfolio level, what are the pressures should we see potentially what we could best describe as a perfect storm of those secondary storms and a quite active um quite active NAT CAT hurricane season?
00:01:18.079 --> 00:01:30.560
Yeah, it's a super interesting uh time, John, as you say, looking back over the last year in particular, because I think historically we've had a reinsurance market that's been quite geared towards hurricanes.
00:01:30.640 --> 00:01:37.519
At the end of the day, you know, hurricanes were a thing that happened once every 10 years or so on a severity that you you would register.
00:01:37.680 --> 00:01:42.719
Uh and if they didn't happen, it was a great year, and if they did happen, it was a bad year.
00:01:42.799 --> 00:01:47.519
And and the investors understood that, the markets understood that, the buyers of reinsurance understood that.
00:01:47.599 --> 00:01:56.159
But as you said, uh last year was a a bit of an exceptional one in that we had the the sort of single largest event was actually a wildfire.
00:01:56.239 --> 00:02:05.519
Uh we had or a group of wildfires happening out in California, uh, driving some 60 billion worth of losses, if you and then of that some 40 billion of insured losses.
00:02:05.599 --> 00:02:17.439
Uh, and then we still got to above 100 billion in overall insured losses from a very large number of uh billion dollar plus individual events that were mostly severe convective storms.
00:02:17.599 --> 00:02:28.879
And all of that despite the fact that there were no hurricanes, which normally you'd think reinsurers amazing news, uh, and also good news for the insurers worried about what might be set within their portfolios.
00:02:29.039 --> 00:02:47.759
Uh so I think the rules of of the how you do reinsurance are having to change a little bit to adapt to not only new uh perils, like you say, like secondary wildfire type perils, but also a lot of other uh pieces that that keep an insurance portfolio manager up at night.
00:02:48.639 --> 00:03:03.280
I mean, the the interesting I suppose is that um in some ways the reinsurance market, uh as with the the broader insurance market, is having to sort of redefine what it deems the new normal.
00:03:03.840 --> 00:03:05.759
Yeah, I I absolutely agree.
00:03:06.080 --> 00:03:16.080
We we've historically, I think, looked at a a year with over a hundred billion of insured losses as a bad year, but it has now very much become the average.
00:03:16.400 --> 00:03:28.159
I and and what's scary, I think, for some of those people now, in particular when you look back at some of the historically devastating hurricanes we have.
00:03:28.319 --> 00:03:37.919
You look back at Katrina, Rita Wilner, the KRW year of 2005, for example, you know, that series of hurricanes alone were more than 100 billion by themselves.
00:03:38.080 --> 00:03:47.199
So resting on everybody's minds is this sort of question: what if we had a KRW on top of a loss year that was already$100 billion by itself?
00:03:47.280 --> 00:03:50.240
Because then we reach you know more than$200 billion in a loss year.
00:03:50.319 --> 00:03:53.439
And that's the whole new normal in terms of what could be extreme.
00:03:53.520 --> 00:04:13.840
And in fact, actually, even uh just a couple of years ago, I remember doing a live podcast uh ourselves on the on the reinsurance side, looking at what would happen if Hurricane Milton, as it were, had a direct hit on Tampa, and in the end it sort of swerved uh at the very last minute and was actually much less impactful.
00:04:13.919 --> 00:04:24.079
But the the estimates that were coming out had that single hurricane hit uh directly where it was uh going into Florida, we we could have been looking at$100 billion from one single event.
00:04:24.319 --> 00:04:31.600
Uh so yeah, the the rules, the the the anchor points that people have had in their minds for a long time are certainly shifting.
00:04:31.759 --> 00:04:53.439
And I think that's driving a bit of a tug of war between sedents and reinsurers that we've seen in the past few years, but it's also driving a need for greater visibility and understanding of what are actually all of the different parts of the portfolio that might be exposed, and what are the protections that are in place to make sure that you are you are proportionately covered in each case.
00:04:54.000 --> 00:04:59.040
Well, obviously, the the market is now blessed with far more data than it ever has been.
00:04:59.199 --> 00:05:29.839
But I suppose in some ways, and I know there's a feeling amongst um some in the market that basically the the mark the the risks have changed so fundamentally that the the historic, you know, the models that are built on historic data are starting to become irrelevant, and that actually in a way the reinsurance market's got to try and use the data it's got at the moment to almost act as a crystal ball to look at where they the the direction of future travel.
00:05:30.639 --> 00:05:32.399
Yeah, that's a great point, John.
00:05:32.639 --> 00:05:51.839
And and and actually one of the the challenges there is that there is perhaps less ability to rely on on historical data not only in traditional categories but also in a lot of the newer categories where these emerging perils are happening.
00:05:52.160 --> 00:06:07.360
So I if we take the traditional perils first, I earthquake's the big one that never gets talked about enough, but obviously we've got this looming possibility of a massive earthquake where it's very, very hard to tell you know whether our historical data is relevant at all to that equation.
00:06:07.600 --> 00:06:20.399
Hurricanes, we thought we had quite a good idea of where that should be, but now there's a bit of a challenge to work out uh if the climate is changing, what does that mean for how our hurricane models should respond?
00:06:20.480 --> 00:06:27.439
Uh and and are we accounting enough for you know the potential impacts of a warming uh climate?
00:06:27.600 --> 00:06:29.759
So that those are sort of the traditional concerns.
00:06:29.839 --> 00:06:45.279
But then on the other hand, on the the sort of more secondary or just non-catastrophe risk perils, you've got a casualty world at the moment that's sort of defying expectations in terms of the social inflation impact that we've seen for many years.
00:06:45.439 --> 00:07:00.720
Uh, you're seeing entirely new perils emerge in the casualty space, uh, where lawsuits are starting to build around things like, for example, uh addictive software design is one that we've seen recently come about with lots of lawsuits there.
00:07:00.800 --> 00:07:06.160
We're seeing things around uh food which is ultra-processed and therefore not really food.
00:07:06.240 --> 00:07:16.480
Uh so everybody's sort of looking around, you know, is there going to be another one of these asbestosis style things that emerges suddenly, and how do we model for the possibility that something like that can happen?
00:07:16.639 --> 00:07:39.680
And then you jump to something like cyber, of course, which is also very, very new, and I know you've done a dedicated episode on that, uh, where yesterday's data feels out of date already because the base of AI is moving things forward so quickly because you're having to overlay on that as well the geopolitical tensions around the world and the impact that might have not only on cyber but also on all sorts of other risks.
00:07:39.759 --> 00:07:52.000
We're seeing big marine risks already take shape outside of conflict zones and only really getting a sense of what they could look like across specialty marine aviation, political violence, etc.
00:07:52.240 --> 00:07:55.680
Uh, we're in the early days there, potentially, of what could happen there as well.
00:07:55.839 --> 00:08:16.720
So I so I I know that we we we we do a lot of work ourselves to support reinsurance buyers uh in particular, but we I don't envy them the job that they've got in terms of trying to work out how do you spread your reinsurance spend between all these different things that could happen and and potentially simultaneously as well, and with only a limited amount of data uh to support your decisions.
00:08:17.040 --> 00:08:25.680
I suppose the other interesting thing about it is uh as you say, there is this growing understanding now, I think, from the of the sheer interconnectedness of risk.
00:08:25.920 --> 00:08:44.159
You know, and and I know that the primary insurance market is something the primary insurance market has been very vocal on in recent months, and they're trying to talk to their they're trying to talk to their major corporate clients to say, look, do you appreciate the sheer, you know, do you appreciate the knock-on effect?
00:08:44.240 --> 00:09:08.399
And I know the World Economic Forum has warned about this in their annual global risk report for 2026, but I suppose in a way, is that a conversation that the reinsurance market is going to have to start having with exceedance in terms of, you know, as you say, we need to understand our exposures, and to do so, we need to be confident that you understand yours.
00:09:08.720 --> 00:09:10.799
I th I I think absolutely, John.
00:09:11.039 --> 00:09:28.159
I think we're at an age now where the world is so interconnected that we almost have to become chaos theorists uh in reinsurance and to make sense of you know the potential knock-on effects that can happen uh from a single source potentially but in so many different dimensions.
00:09:28.240 --> 00:09:45.120
I mean, if you were to take the change in the US administration uh as an example, uh the impacts of the recent uh intervention, shall we say, in Iran, for example, uh are going to be felt deeply economically.
00:09:45.279 --> 00:09:52.399
I, as we all know, watching the sort of potential fuel crises and inflation risks and uncertainty that stems from that.
00:09:52.639 --> 00:10:05.600
I that in itself, even if you've got nothing to do with the Middle East or or any of the um the actual exposures on the ground or nearby there, you're going to be impacted as somebody who manages assets.
00:10:05.679 --> 00:10:11.120
If you're an insurance company or a reinsurer, you know, you you have a lot of assets that will be affected by that uh indirectly.
00:10:11.279 --> 00:10:15.840
There's also then all the stuff that it affects quite directly on the ground, if that's part of your portfolio.
00:10:16.080 --> 00:10:31.919
I there's also, again, this is something that came up quite a lot in uh the reinsurance conferences going back a couple of years ago, but uh the possibility for political violence and unrest that's occurred as a result of a change of administration in the past as well.
00:10:32.000 --> 00:10:36.720
You know, we we had the the storming of the capital, etc., a few years ago that we saw.
00:10:36.879 --> 00:10:41.600
All of these things can come from potentially just one person uh taking office.
00:10:41.679 --> 00:10:47.440
So we we've we've seen a huge amount of disruption caused uh from one butterfly.
00:10:47.519 --> 00:10:54.559
Uh, and I think reinsurers need to be aware of that and try and spot the independency, interdependencies in their portfolios.
00:10:54.639 --> 00:11:15.120
And similarly, sedents need to be able to look at all the different exposures they have and the risk that a given scenario doesn't affect just one of their programs, but potentially several of them, and the order in which they have their different reinsurance programs anuring to each other could quite materially affect what they're able to recover in a sort of chain reaction of events.
00:11:15.279 --> 00:11:21.279
I've not even touched on you know where we could end up as all these these events bleed into each other.
00:11:22.159 --> 00:11:23.840
Yeah, because it is quite interesting.
00:11:23.919 --> 00:11:28.720
I I think in a way, um, I'm not going to show my age here now, uh Ben.
00:11:29.120 --> 00:11:33.600
Um I think in a way the the whole viewpoint of the market has changed.
00:11:33.919 --> 00:11:46.240
26 years ago, when I I have first arrived as a journalist in the market, um I'm you know, obviously trying to get get an idea of of sort of the thinking around the market.
00:11:46.399 --> 00:11:54.399
I I remember talking to a couple of Lloyd's um underwriters, and I asked them, I said, Well, what is your worst case scenario?
00:11:54.480 --> 00:12:21.600
And as I said, this was this was 1999, 2000, and their response was and how uh how you know sort of how prophetic this was two 747s colliding over Manhattan, and that was their worst case scenario, and and obviously, you know, fate 11 years ago 11 years later, we saw you know, we saw something like that.
00:12:21.759 --> 00:12:33.039
But I suppose the issue is then, if I was to ask that question now, I don't think we'd be in a situation where it would be, well, actually, we're looking at a single event.
00:12:33.200 --> 00:12:40.879
Well, I think now may well be the situation where they're saying, well, to be quite honest, it would be that that perfect storm.
00:12:41.200 --> 00:12:57.200
Yeah, uh, I think you you raise a really, really good point there with uh, you know, what used to be effectively the and and still is actually quite similar, the the Lloyd's realistic disaster scenarios, for example, where they do measure you know what their exposure would be to events like if two planes collide over Manhattan.
00:12:57.360 --> 00:13:06.399
Um, and I think absolutely right to question whether they consider just how much those events can cause a series of knock-on effects.
00:13:06.480 --> 00:13:15.039
So an example that jumped to mind whilst you were talking in terms of how we might need to evolve our thinking is the the rise of the data centers that we're seeing at the moment.
00:13:15.200 --> 00:13:35.840
I, you know, there's I I think at the current number, so depending on how many weeks it's been, if you listen to this podcast when it first comes out or in a few weeks after that, currently we're on something like 600 billion committed towards new data center construction for this year, and that's expected to keep on going up as the Magnificent 7 and others keep putting money into these data centers.
00:13:36.080 --> 00:13:46.240
I as a as an asset, first of all, there's the potential for a realistic disaster scenario exclusively on that asset being affected in some way.
00:13:46.320 --> 00:14:03.039
So let's say, you know, there is a fire at a dentist data center and the data center is destroyed, or a tornado uh passes through, as lots of them seem to be in a tornado path, or civilian unrest because everyone's very angry about their bills going up, causes somebody to try and you know burn down one of these data centers, whatever it might be.
00:14:03.279 --> 00:14:26.799
Uh the knock-on effects of that also could easily carry on through to other areas in the sense of uh, for example, if that means an outage of some kind that causes uh unrest in turn and causes lots of uh unavailability of systems, as we increasingly put critical systems dependent on uh automated or or AI-managed uh sort of interfaces.
00:14:26.879 --> 00:14:49.600
So we are as much as we uh are growing our awareness of the interdependency of lots of different risks, we're actually actually constantly increasing the interdependency uh all the time by at the moment putting more and more things into the hand of computer systems and more and more operations uh having dependency built around the availability of data centers.
00:14:49.759 --> 00:15:06.000
I I think you'll probably remember as as well, John, the the jokes uh that followed the Cloudflare outage not long ago, or something where you know people couldn't get into their houses all of a sudden because they had smart locks uh on their apartments uh and and they didn't work anymore because of an outage.
00:15:06.159 --> 00:15:15.840
Uh that's the type of knock-on that we can find in a very connected digital as well as physical world that might really start to change the way we think about risk.
00:15:16.240 --> 00:15:17.919
I suppose that that is the issue, isn't it?
00:15:18.000 --> 00:15:25.919
As you say, I think where we are now is that the market is having to fundamentally change the way in which it thinks about risk.
00:15:26.080 --> 00:15:30.240
Um not I suppose not the approach, because obviously prudence is always going to be there.
00:15:30.320 --> 00:15:51.919
And I suppose the underlying issue about actually you know pricing accurately, understanding your exposures, and therefore factoring that into your risk appetite is there, but I suppose in a way it's it's it's the risks themselves and how how that how they actually approach that.
00:15:52.000 --> 00:16:24.320
Because of course the other issue is what we are seeing, and and you you know, and I think we in a number of the areas that we've talked about over the last 15 minutes, you know, there is a growing there is this growing protection gap where you know there's a now a disconnect between the the demands or the uh the demands of the the client and we're talking about the the uh the risk manager, the corporate risk manager, and the ability of the and the ability of the um the primary market to deliver that capacity.
00:16:24.399 --> 00:16:30.399
I mean clearly that's that's a debate over you know it's a debate over um over cyber, certainly.
00:16:30.559 --> 00:16:35.279
Um and again, you know, certain natural, you know, we saw it with the reinsurers.
00:16:35.360 --> 00:16:41.919
They had a very conservative view a few years ago over natural natural caps after after some pretty tough years.
00:16:42.159 --> 00:16:50.240
So I suppose in a way, but again, it's gonna come back on the reinsurers because of course there'll be pressure on the seedants to say, look, we need the capacity.
00:16:50.399 --> 00:16:53.519
Well, we've only got a certain amount of risk.
00:16:53.679 --> 00:17:17.839
What we want to do is then is obviously offset mitigate that risk by increasing our reinsurance capacity, which again is going to put the you know, I mean we'll forget the retro market, but I mean on on the on the reinsurers, I think we can fairly say they're gonna be coming under a bit more pressure in the next you know three to five years to start committing capacity to to risks that were never really on their radar ten years ago.
00:17:18.160 --> 00:17:20.000
Yeah, I I I totally agree, actually.
00:17:20.240 --> 00:17:35.200
I think there's a a couple of really interesting factors at play here where on one hand, as you say, you're in a a world that's suddenly started, it feels to change a lot faster than it was changing before.
00:17:35.519 --> 00:17:43.680
And that is putting a challenge on what we used to think of, I suppose, as a rinse and rin repeat renewal process.
00:17:43.839 --> 00:17:51.759
You know, historically, a an insurance company would have a pretty similar-looking portfolio in one year compared to the previous one.
00:17:51.839 --> 00:17:56.240
And so when it came to buying reinsurance, they would buy a pretty similar-looking reinsurance program.
00:17:56.559 --> 00:18:06.960
And depending on if it was a soft market or a hard market, you know, they'd be pushing for different outcomes in terms of the price that they'd pay for that coverage, but they'd be buying pretty much the same thing.
00:18:07.119 --> 00:18:33.440
Uh whereas now, actually, suddenly we're in a position where, due to that protection gap that you highlighted, and due to you know changing demands from the end, consumers of insurance, business or or personal or other, the the need is actually for the reinsurers to innovate their products and to be able to offer different kinds of coverages, uh different peril uh coverages as well, such that the insurance companies are able to support their clients in turn.
00:18:33.680 --> 00:19:18.160
And I think actually there was a bit of a signal uh this renewal season just gone, so so 1.1 2026 in particular, that we're failing a little bit there at the moment as the reinsurance market, because despite the best efforts of the brokers, uh when the reinsurance buyers were able to realize quite a lot of savings uh on their reinsurance, like traditional purchases that they would normally make, they seemed by and large to choose not to invest the savings uh that they'd made in additional coverage in other areas, uh, despite the brokers saying, you know, we really should try and, you know, you've saved 30 million or whatever because prices have come down 12% or whatever it might be.
00:19:18.319 --> 00:19:23.599
Shall we spend that 30 million on coverage for this this other peril that we know is in your portfolio?
00:19:23.759 --> 00:19:29.839
And we didn't manage as an industry to find a match in the majority of cases there where they would redeploy that money.
00:19:29.920 --> 00:19:34.559
And Sedents, by and large, chose that it was better to retain that extra capital.
00:19:34.720 --> 00:19:46.319
So I think there's work to be done uh between the buyers, the brokers, and the reinsurers to figure out okay, how can we cover those additional areas that we've we've not quite covered in the portfolio.
00:19:46.480 --> 00:20:01.759
Uh and I think there's additional pressures from others who will and others who can provide solutions, including the ILS uh market at the moment, which is ever more active uh both on the traditional front but also on some of these non-traditional areas as well.
00:20:02.240 --> 00:20:02.960
I think you're right.
00:20:03.039 --> 00:20:18.400
I mean the thing is Ben, um, as you say, uh I I think what the what the past 20 minutes has has highlighted is the the sheer breadth of the the new the new risks, the new threats, the new dynamics that the market is facing.
00:20:18.559 --> 00:20:30.400
Um I think you know uh there'll there's there is and I'm sure there will continue to be a great deal of debate in the uh in the underwriting rooms across the world on on how best to approach this.
00:20:30.480 --> 00:20:33.039
But as I said, I can only thank you for your insight.
00:20:33.200 --> 00:20:35.359
Thank you for your expertise today.
00:20:35.519 --> 00:20:48.640
Um and and hopefully, as I said, um maybe in a in sort of six six months' time we can we can get back together and see how how how how well the reinsurance market has uh has heeded the warnings.
00:20:49.279 --> 00:20:49.680
Indeed.
00:20:49.839 --> 00:20:50.319
Thank you, John.
00:20:50.400 --> 00:20:54.160
It's been a real pleasure and uh look forward to listening and talking again soon.
00:20:54.720 --> 00:21:02.720
Thank you for listening to Maximum Exposure Talking Emerging Risks, a new series of podcasts brought to you by emergingrisks.co.uk.
00:21:03.839 --> 00:21:14.640
We are a dedicated publication looking at the future of risk, delivering insightful news and cogent analysis on the key emerging topics facing the market in these turbulent times.
00:21:15.440 --> 00:21:30.079
With over 150,000 impressions every month, Emerging Risks has now established itself as the go to read for all those concerned with the latest analysis and trends in a dynamic and fast moving risk environment.
00:00:07.200 --> 00:00:12.720
Welcome to the latest edition of the Maximum Exposure Podcast, brought to you by Emerging Risks.
00:00:13.119 --> 00:00:25.440
Ben Rose, president and co-founder of Supersed, discusses with John Guy why re-insurers are having to reassess their approach to risk as exposures become ever more interconnected.
00:00:27.839 --> 00:00:37.359
I'm delighted to welcome Ben Rose, president and co-founder of Supersed, with me on today's edition of the Maximum Exposure podcast brought to you by Emerging Risk.
00:00:37.920 --> 00:00:40.880
Ben, as I said, welcome to the podcast.
00:00:41.119 --> 00:00:43.920
It's been a really interesting year for 2025.
00:00:44.159 --> 00:00:57.200
We saw a on the natural peril side, certainly, we saw the rise of uh secondary perils, but fortunately, I think for the insurance market, we had a very subpart North Atlantic hurricane season.
00:00:57.359 --> 00:01:00.799
However, I suppose the industry cannot really rest on its laurels.
00:01:00.880 --> 00:01:17.760
I mean, from your point of view, on a portfolio level, what are the pressures should we see potentially what we could best describe as a perfect storm of those secondary storms and a quite active um quite active NAT CAT hurricane season?
00:01:18.079 --> 00:01:30.560
Yeah, it's a super interesting uh time, John, as you say, looking back over the last year in particular, because I think historically we've had a reinsurance market that's been quite geared towards hurricanes.
00:01:30.640 --> 00:01:37.519
At the end of the day, you know, hurricanes were a thing that happened once every 10 years or so on a severity that you you would register.
00:01:37.680 --> 00:01:42.719
Uh and if they didn't happen, it was a great year, and if they did happen, it was a bad year.
00:01:42.799 --> 00:01:47.519
And and the investors understood that, the markets understood that, the buyers of reinsurance understood that.
00:01:47.599 --> 00:01:56.159
But as you said, uh last year was a a bit of an exceptional one in that we had the the sort of single largest event was actually a wildfire.
00:01:56.239 --> 00:02:05.519
Uh we had or a group of wildfires happening out in California, uh, driving some 60 billion worth of losses, if you and then of that some 40 billion of insured losses.
00:02:05.599 --> 00:02:17.439
Uh, and then we still got to above 100 billion in overall insured losses from a very large number of uh billion dollar plus individual events that were mostly severe convective storms.
00:02:17.599 --> 00:02:28.879
And all of that despite the fact that there were no hurricanes, which normally you'd think reinsurers amazing news, uh, and also good news for the insurers worried about what might be set within their portfolios.
00:02:29.039 --> 00:02:47.759
Uh so I think the rules of of the how you do reinsurance are having to change a little bit to adapt to not only new uh perils, like you say, like secondary wildfire type perils, but also a lot of other uh pieces that that keep an insurance portfolio manager up at night.
00:02:48.639 --> 00:03:03.280
I mean, the the interesting I suppose is that um in some ways the reinsurance market, uh as with the the broader insurance market, is having to sort of redefine what it deems the new normal.
00:03:03.840 --> 00:03:05.759
Yeah, I I absolutely agree.
00:03:06.080 --> 00:03:16.080
We we've historically, I think, looked at a a year with over a hundred billion of insured losses as a bad year, but it has now very much become the average.
00:03:16.400 --> 00:03:28.159
I and and what's scary, I think, for some of those people now, in particular when you look back at some of the historically devastating hurricanes we have.
00:03:28.319 --> 00:03:37.919
You look back at Katrina, Rita Wilner, the KRW year of 2005, for example, you know, that series of hurricanes alone were more than 100 billion by themselves.
00:03:38.080 --> 00:03:47.199
So resting on everybody's minds is this sort of question: what if we had a KRW on top of a loss year that was already$100 billion by itself?
00:03:47.280 --> 00:03:50.240
Because then we reach you know more than$200 billion in a loss year.
00:03:50.319 --> 00:03:53.439
And that's the whole new normal in terms of what could be extreme.
00:03:53.520 --> 00:04:13.840
And in fact, actually, even uh just a couple of years ago, I remember doing a live podcast uh ourselves on the on the reinsurance side, looking at what would happen if Hurricane Milton, as it were, had a direct hit on Tampa, and in the end it sort of swerved uh at the very last minute and was actually much less impactful.
00:04:13.919 --> 00:04:24.079
But the the estimates that were coming out had that single hurricane hit uh directly where it was uh going into Florida, we we could have been looking at$100 billion from one single event.
00:04:24.319 --> 00:04:31.600
Uh so yeah, the the rules, the the the anchor points that people have had in their minds for a long time are certainly shifting.
00:04:31.759 --> 00:04:53.439
And I think that's driving a bit of a tug of war between sedents and reinsurers that we've seen in the past few years, but it's also driving a need for greater visibility and understanding of what are actually all of the different parts of the portfolio that might be exposed, and what are the protections that are in place to make sure that you are you are proportionately covered in each case.
00:04:54.000 --> 00:04:59.040
Well, obviously, the the market is now blessed with far more data than it ever has been.
00:04:59.199 --> 00:05:29.839
But I suppose in some ways, and I know there's a feeling amongst um some in the market that basically the the mark the the risks have changed so fundamentally that the the historic, you know, the models that are built on historic data are starting to become irrelevant, and that actually in a way the reinsurance market's got to try and use the data it's got at the moment to almost act as a crystal ball to look at where they the the direction of future travel.
00:05:30.639 --> 00:05:32.399
Yeah, that's a great point, John.
00:05:32.639 --> 00:05:51.839
And and and actually one of the the challenges there is that there is perhaps less ability to rely on on historical data not only in traditional categories but also in a lot of the newer categories where these emerging perils are happening.
00:05:52.160 --> 00:06:07.360
So I if we take the traditional perils first, I earthquake's the big one that never gets talked about enough, but obviously we've got this looming possibility of a massive earthquake where it's very, very hard to tell you know whether our historical data is relevant at all to that equation.
00:06:07.600 --> 00:06:20.399
Hurricanes, we thought we had quite a good idea of where that should be, but now there's a bit of a challenge to work out uh if the climate is changing, what does that mean for how our hurricane models should respond?
00:06:20.480 --> 00:06:27.439
Uh and and are we accounting enough for you know the potential impacts of a warming uh climate?
00:06:27.600 --> 00:06:29.759
So that those are sort of the traditional concerns.
00:06:29.839 --> 00:06:45.279
But then on the other hand, on the the sort of more secondary or just non-catastrophe risk perils, you've got a casualty world at the moment that's sort of defying expectations in terms of the social inflation impact that we've seen for many years.
00:06:45.439 --> 00:07:00.720
Uh, you're seeing entirely new perils emerge in the casualty space, uh, where lawsuits are starting to build around things like, for example, uh addictive software design is one that we've seen recently come about with lots of lawsuits there.
00:07:00.800 --> 00:07:06.160
We're seeing things around uh food which is ultra-processed and therefore not really food.
00:07:06.240 --> 00:07:16.480
Uh so everybody's sort of looking around, you know, is there going to be another one of these asbestosis style things that emerges suddenly, and how do we model for the possibility that something like that can happen?
00:07:16.639 --> 00:07:39.680
And then you jump to something like cyber, of course, which is also very, very new, and I know you've done a dedicated episode on that, uh, where yesterday's data feels out of date already because the base of AI is moving things forward so quickly because you're having to overlay on that as well the geopolitical tensions around the world and the impact that might have not only on cyber but also on all sorts of other risks.
00:07:39.759 --> 00:07:52.000
We're seeing big marine risks already take shape outside of conflict zones and only really getting a sense of what they could look like across specialty marine aviation, political violence, etc.
00:07:52.240 --> 00:07:55.680
Uh, we're in the early days there, potentially, of what could happen there as well.
00:07:55.839 --> 00:08:16.720
So I so I I know that we we we we do a lot of work ourselves to support reinsurance buyers uh in particular, but we I don't envy them the job that they've got in terms of trying to work out how do you spread your reinsurance spend between all these different things that could happen and and potentially simultaneously as well, and with only a limited amount of data uh to support your decisions.
00:08:17.040 --> 00:08:25.680
I suppose the other interesting thing about it is uh as you say, there is this growing understanding now, I think, from the of the sheer interconnectedness of risk.
00:08:25.920 --> 00:08:44.159
You know, and and I know that the primary insurance market is something the primary insurance market has been very vocal on in recent months, and they're trying to talk to their they're trying to talk to their major corporate clients to say, look, do you appreciate the sheer, you know, do you appreciate the knock-on effect?
00:08:44.240 --> 00:09:08.399
And I know the World Economic Forum has warned about this in their annual global risk report for 2026, but I suppose in a way, is that a conversation that the reinsurance market is going to have to start having with exceedance in terms of, you know, as you say, we need to understand our exposures, and to do so, we need to be confident that you understand yours.
00:09:08.720 --> 00:09:10.799
I th I I think absolutely, John.
00:09:11.039 --> 00:09:28.159
I think we're at an age now where the world is so interconnected that we almost have to become chaos theorists uh in reinsurance and to make sense of you know the potential knock-on effects that can happen uh from a single source potentially but in so many different dimensions.
00:09:28.240 --> 00:09:45.120
I mean, if you were to take the change in the US administration uh as an example, uh the impacts of the recent uh intervention, shall we say, in Iran, for example, uh are going to be felt deeply economically.
00:09:45.279 --> 00:09:52.399
I, as we all know, watching the sort of potential fuel crises and inflation risks and uncertainty that stems from that.
00:09:52.639 --> 00:10:05.600
I that in itself, even if you've got nothing to do with the Middle East or or any of the um the actual exposures on the ground or nearby there, you're going to be impacted as somebody who manages assets.
00:10:05.679 --> 00:10:11.120
If you're an insurance company or a reinsurer, you know, you you have a lot of assets that will be affected by that uh indirectly.
00:10:11.279 --> 00:10:15.840
There's also then all the stuff that it affects quite directly on the ground, if that's part of your portfolio.
00:10:16.080 --> 00:10:31.919
I there's also, again, this is something that came up quite a lot in uh the reinsurance conferences going back a couple of years ago, but uh the possibility for political violence and unrest that's occurred as a result of a change of administration in the past as well.
00:10:32.000 --> 00:10:36.720
You know, we we had the the storming of the capital, etc., a few years ago that we saw.
00:10:36.879 --> 00:10:41.600
All of these things can come from potentially just one person uh taking office.
00:10:41.679 --> 00:10:47.440
So we we've we've seen a huge amount of disruption caused uh from one butterfly.
00:10:47.519 --> 00:10:54.559
Uh, and I think reinsurers need to be aware of that and try and spot the independency, interdependencies in their portfolios.
00:10:54.639 --> 00:11:15.120
And similarly, sedents need to be able to look at all the different exposures they have and the risk that a given scenario doesn't affect just one of their programs, but potentially several of them, and the order in which they have their different reinsurance programs anuring to each other could quite materially affect what they're able to recover in a sort of chain reaction of events.
00:11:15.279 --> 00:11:21.279
I've not even touched on you know where we could end up as all these these events bleed into each other.
00:11:22.159 --> 00:11:23.840
Yeah, because it is quite interesting.
00:11:23.919 --> 00:11:28.720
I I think in a way, um, I'm not going to show my age here now, uh Ben.
00:11:29.120 --> 00:11:33.600
Um I think in a way the the whole viewpoint of the market has changed.
00:11:33.919 --> 00:11:46.240
26 years ago, when I I have first arrived as a journalist in the market, um I'm you know, obviously trying to get get an idea of of sort of the thinking around the market.
00:11:46.399 --> 00:11:54.399
I I remember talking to a couple of Lloyd's um underwriters, and I asked them, I said, Well, what is your worst case scenario?
00:11:54.480 --> 00:12:21.600
And as I said, this was this was 1999, 2000, and their response was and how uh how you know sort of how prophetic this was two 747s colliding over Manhattan, and that was their worst case scenario, and and obviously, you know, fate 11 years ago 11 years later, we saw you know, we saw something like that.
00:12:21.759 --> 00:12:33.039
But I suppose the issue is then, if I was to ask that question now, I don't think we'd be in a situation where it would be, well, actually, we're looking at a single event.
00:12:33.200 --> 00:12:40.879
Well, I think now may well be the situation where they're saying, well, to be quite honest, it would be that that perfect storm.
00:12:41.200 --> 00:12:57.200
Yeah, uh, I think you you raise a really, really good point there with uh, you know, what used to be effectively the and and still is actually quite similar, the the Lloyd's realistic disaster scenarios, for example, where they do measure you know what their exposure would be to events like if two planes collide over Manhattan.
00:12:57.360 --> 00:13:06.399
Um, and I think absolutely right to question whether they consider just how much those events can cause a series of knock-on effects.
00:13:06.480 --> 00:13:15.039
So an example that jumped to mind whilst you were talking in terms of how we might need to evolve our thinking is the the rise of the data centers that we're seeing at the moment.
00:13:15.200 --> 00:13:35.840
I, you know, there's I I think at the current number, so depending on how many weeks it's been, if you listen to this podcast when it first comes out or in a few weeks after that, currently we're on something like 600 billion committed towards new data center construction for this year, and that's expected to keep on going up as the Magnificent 7 and others keep putting money into these data centers.
00:13:36.080 --> 00:13:46.240
I as a as an asset, first of all, there's the potential for a realistic disaster scenario exclusively on that asset being affected in some way.
00:13:46.320 --> 00:14:03.039
So let's say, you know, there is a fire at a dentist data center and the data center is destroyed, or a tornado uh passes through, as lots of them seem to be in a tornado path, or civilian unrest because everyone's very angry about their bills going up, causes somebody to try and you know burn down one of these data centers, whatever it might be.
00:14:03.279 --> 00:14:26.799
Uh the knock-on effects of that also could easily carry on through to other areas in the sense of uh, for example, if that means an outage of some kind that causes uh unrest in turn and causes lots of uh unavailability of systems, as we increasingly put critical systems dependent on uh automated or or AI-managed uh sort of interfaces.
00:14:26.879 --> 00:14:49.600
So we are as much as we uh are growing our awareness of the interdependency of lots of different risks, we're actually actually constantly increasing the interdependency uh all the time by at the moment putting more and more things into the hand of computer systems and more and more operations uh having dependency built around the availability of data centers.
00:14:49.759 --> 00:15:06.000
I I think you'll probably remember as as well, John, the the jokes uh that followed the Cloudflare outage not long ago, or something where you know people couldn't get into their houses all of a sudden because they had smart locks uh on their apartments uh and and they didn't work anymore because of an outage.
00:15:06.159 --> 00:15:15.840
Uh that's the type of knock-on that we can find in a very connected digital as well as physical world that might really start to change the way we think about risk.
00:15:16.240 --> 00:15:17.919
I suppose that that is the issue, isn't it?
00:15:18.000 --> 00:15:25.919
As you say, I think where we are now is that the market is having to fundamentally change the way in which it thinks about risk.
00:15:26.080 --> 00:15:30.240
Um not I suppose not the approach, because obviously prudence is always going to be there.
00:15:30.320 --> 00:15:51.919
And I suppose the underlying issue about actually you know pricing accurately, understanding your exposures, and therefore factoring that into your risk appetite is there, but I suppose in a way it's it's it's the risks themselves and how how that how they actually approach that.
00:15:52.000 --> 00:16:24.320
Because of course the other issue is what we are seeing, and and you you know, and I think we in a number of the areas that we've talked about over the last 15 minutes, you know, there is a growing there is this growing protection gap where you know there's a now a disconnect between the the demands or the uh the demands of the the client and we're talking about the the uh the risk manager, the corporate risk manager, and the ability of the and the ability of the um the primary market to deliver that capacity.
00:16:24.399 --> 00:16:30.399
I mean clearly that's that's a debate over you know it's a debate over um over cyber, certainly.
00:16:30.559 --> 00:16:35.279
Um and again, you know, certain natural, you know, we saw it with the reinsurers.
00:16:35.360 --> 00:16:41.919
They had a very conservative view a few years ago over natural natural caps after after some pretty tough years.
00:16:42.159 --> 00:16:50.240
So I suppose in a way, but again, it's gonna come back on the reinsurers because of course there'll be pressure on the seedants to say, look, we need the capacity.
00:16:50.399 --> 00:16:53.519
Well, we've only got a certain amount of risk.
00:16:53.679 --> 00:17:17.839
What we want to do is then is obviously offset mitigate that risk by increasing our reinsurance capacity, which again is going to put the you know, I mean we'll forget the retro market, but I mean on on the on the reinsurers, I think we can fairly say they're gonna be coming under a bit more pressure in the next you know three to five years to start committing capacity to to risks that were never really on their radar ten years ago.
00:17:18.160 --> 00:17:20.000
Yeah, I I I totally agree, actually.
00:17:20.240 --> 00:17:35.200
I think there's a a couple of really interesting factors at play here where on one hand, as you say, you're in a a world that's suddenly started, it feels to change a lot faster than it was changing before.
00:17:35.519 --> 00:17:43.680
And that is putting a challenge on what we used to think of, I suppose, as a rinse and rin repeat renewal process.
00:17:43.839 --> 00:17:51.759
You know, historically, a an insurance company would have a pretty similar-looking portfolio in one year compared to the previous one.
00:17:51.839 --> 00:17:56.240
And so when it came to buying reinsurance, they would buy a pretty similar-looking reinsurance program.
00:17:56.559 --> 00:18:06.960
And depending on if it was a soft market or a hard market, you know, they'd be pushing for different outcomes in terms of the price that they'd pay for that coverage, but they'd be buying pretty much the same thing.
00:18:07.119 --> 00:18:33.440
Uh whereas now, actually, suddenly we're in a position where, due to that protection gap that you highlighted, and due to you know changing demands from the end, consumers of insurance, business or or personal or other, the the need is actually for the reinsurers to innovate their products and to be able to offer different kinds of coverages, uh different peril uh coverages as well, such that the insurance companies are able to support their clients in turn.
00:18:33.680 --> 00:19:18.160
And I think actually there was a bit of a signal uh this renewal season just gone, so so 1.1 2026 in particular, that we're failing a little bit there at the moment as the reinsurance market, because despite the best efforts of the brokers, uh when the reinsurance buyers were able to realize quite a lot of savings uh on their reinsurance, like traditional purchases that they would normally make, they seemed by and large to choose not to invest the savings uh that they'd made in additional coverage in other areas, uh, despite the brokers saying, you know, we really should try and, you know, you've saved 30 million or whatever because prices have come down 12% or whatever it might be.
00:19:18.319 --> 00:19:23.599
Shall we spend that 30 million on coverage for this this other peril that we know is in your portfolio?
00:19:23.759 --> 00:19:29.839
And we didn't manage as an industry to find a match in the majority of cases there where they would redeploy that money.
00:19:29.920 --> 00:19:34.559
And Sedents, by and large, chose that it was better to retain that extra capital.
00:19:34.720 --> 00:19:46.319
So I think there's work to be done uh between the buyers, the brokers, and the reinsurers to figure out okay, how can we cover those additional areas that we've we've not quite covered in the portfolio.
00:19:46.480 --> 00:20:01.759
Uh and I think there's additional pressures from others who will and others who can provide solutions, including the ILS uh market at the moment, which is ever more active uh both on the traditional front but also on some of these non-traditional areas as well.
00:20:02.240 --> 00:20:02.960
I think you're right.
00:20:03.039 --> 00:20:18.400
I mean the thing is Ben, um, as you say, uh I I think what the what the past 20 minutes has has highlighted is the the sheer breadth of the the new the new risks, the new threats, the new dynamics that the market is facing.
00:20:18.559 --> 00:20:30.400
Um I think you know uh there'll there's there is and I'm sure there will continue to be a great deal of debate in the uh in the underwriting rooms across the world on on how best to approach this.
00:20:30.480 --> 00:20:33.039
But as I said, I can only thank you for your insight.
00:20:33.200 --> 00:20:35.359
Thank you for your expertise today.
00:20:35.519 --> 00:20:48.640
Um and and hopefully, as I said, um maybe in a in sort of six six months' time we can we can get back together and see how how how how well the reinsurance market has uh has heeded the warnings.
00:20:49.279 --> 00:20:49.680
Indeed.
00:20:49.839 --> 00:20:50.319
Thank you, John.
00:20:50.400 --> 00:20:54.160
It's been a real pleasure and uh look forward to listening and talking again soon.
00:20:54.720 --> 00:21:02.720
Thank you for listening to Maximum Exposure Talking Emerging Risks, a new series of podcasts brought to you by emergingrisks.co.uk.
00:21:03.839 --> 00:21:14.640
We are a dedicated publication looking at the future of risk, delivering insightful news and cogent analysis on the key emerging topics facing the market in these turbulent times.
00:21:15.440 --> 00:21:30.079
With over 150,000 impressions every month, Emerging Risks has now established itself as the go to read for all those concerned with the latest analysis and trends in a dynamic and fast moving risk environment.
