Fathom's Andrew Smith on flood modelling and closing the protection gap, plus news and views

Hello and welcome to The Green Line, the podcast that tells and do the story shaping each of the specifics we insurance last year, recorded here in the newsrooms of information news in Hong Kong. I'm Mithin Markey, Editor-in-Chief, and in this episode we dive into Zurich's bit for beastly, explore the aftermath of the Victorian bushfires, and assist the impact of other care events in Australia and New Zealand. We'll also discuss the growing significance of data centers in the insurance market in Asia Pacific, and as always, we'll also bring you the stories that we currently investigated. Our guest this week is Andrew Schmidt, co-founder and CEO of Swiss Rehond, flood modeling specialist Fabian. Andrew will share insights on the challenges and closing the flood protection gap, particularly in light of flooding events in China and South and Southeast Asia in 2025. But really the idea for the company Fabian came from our ties with the insurance market, so at the time there were many big floods happening, actually principally in Asia, that kind of woke up the whole global insurance market to the fact that for most of the world we didn't have coverage. This episode I'm joined by the entire Iron Aeditorial team. With me is Deputy Editor Andrew Mullen, Senior Correspondent Aiden Gregory, and Reporters Joanna Nuen and Roshan Ambiar. Welcome everyone. Nice to have you back, Minton. Let's start with the story that's been shaking up global insurance markets. Zürich's $10 billion bid for specialty insurer Beasley. On Monday, Zürich revealed the Swiss carrier has made a enhanced offer for the Lloyd Specialty carrier after the Beasley board rejected multiple such offers in the past. Aiden, can you bring us up to speed on the development surrounding this bid and what's next for World Tights? Sure, so I mean this is big news for the London market and it will also have knock on consequences for Asia too, particularly Singapore. So Zürich has finally decided after trying to woo Beasley's board for the best part of a year that they've gone public with this knock out bid for the company. It's a huge premium, it's a 56% premium to the closing price of Beasley Shares on Friday, which was the last trading day before all of this, all of this was made public. It makes it quite, this bid is quite difficult for Beasley's shoulders to ignore. I guess you could call it a Godfather-like offer. It's one that's very hard to refuse. Whether they still hold out even longer for any even higher bid remains to be seen, but it's very clear that Zürich's very serious about buying the company. It reminds me a little bit of a Viva's takeover of Direct Line in December 2024, although Beasley's an all-skeleton, more exciting business than Direct Line was before it was swallowed up by a Viva. Zürich has been vocal for quite some time now about its ambitions to grow in specialty lines and it also lacks the large Lloyds platform. By buying Beasley, it allows it to achieve both of those objectives and in one fell swoop. Beasley's biggest lines are on property and specialty, but it also has a very fast growing and profitable cyber insurance business, which has been very much at the core of the episode story in the last few years. It's a very exciting prospect and it's fairly clear that the logic behind why Zürich is being very aggressive with its heavily-sweetened offer. Indeed, it does seem like a fairly generous valuation from Zürich and its sort of chimes, but Zürich stated plans to build out its specialty lines, businesses, and that's kind of what all the companies are doing at the moment. But Edan, interestingly, you brought up the Direct Line deal. Can you tell me what the paddles are? Yeah, it does remind me a little bit of that transaction, because in that deal, you also had a very motivated buyer, which had a very clear idea of what it wanted to do, and a Viva had to sweeten its offer before it was finally successful. The final Viva's final bid for Direct Line was done at a 73.3% premium, which is even higher than what Zürich is offering to pay for Beasley. It's pretty clear when you get into this kind of territory that you do force channels to engage, even if they haven't been sort of previously dismissive of the offer. True, I was reading an analyst report which said the offer value is basically at about 2.1 times net asset value of FI-25, which is a significant premium to this historic average M&M multiple inloids of about 1.7 times any V. So it's certainly a great deal for Beasley shareholders, but then there are people who argue that Beasley is a stronger franchisey, and that has a very strong cyber portfolio, which means they're probably holding out for better offers. But what's clear is that, we've been talking about this since last year, we've been talking about acquisitions. We saw that at 1100, there's quite a lot of capital available, this capital could not be deployed, so we definitely see more deals in the coming days, and it's certainly Zürich's offer is just a sign of things to come. So turning to recent cat events, Australia is grappling with deadly bushfires in Victoria, where flames have been raging for weeks, and I've already claimed at least one life. This event evokes memories of the Black Summer a few years ago and draws battles with last year's LA wildfires. Andrew, what's the latest on this situation? Thanks, Matt. You mentioned the wildfires, and I think it is remarkable that has only been one human fatality so far. I think testament to the, I guess, that the people being prepared and the warnings that went out and the warning systems in place. But this this this event is as being massive. It's it's around 4200 hectares of being burnt, and there's been a lot of comparisons with this, and that's that's for us in our region. That's relatively the five times the size of Singapore. There's also been 1,500 structures destroyed, including 435 homes, and on top of that 38,000 livestock have been lost, there's a big big farming community down there, and on top of that there's actually been seven seven racehorses perished. There's a big farm stable down there, and one of the big big trainers lost seven of his horses, who were down there resting, but interesting, they did manage to rescue over 300. So they didn't do too bad getting the horses out. The trainer was actually based in Hong Kong, David Hayes, and he had to rush back to help his sons, you know, with the defences and rescue the horses. So there's a loss there for the race and community, but I think they did quite well to get the majority of the horses out. Insurance Council Australia confirmed that there's been over 2,300 claims lodged across property commercial and motor, and obviously this is this is going to continue to rise, is this event is still ongoing now. A lot of these fires have been brought into control, and there's a lot less than there was, but there is fires still burning, and the impact is going to be felt for a while, and it's definitely some are going to monitor in terms of the losses there, and how much is this event you're going to cost. True, it's just the, it's just a start of the summer, and there's a couple of months to go. To be honest, the last couple of years have been fairly benign for Australia. We haven't seen major catavans, there have been few of course, but one thing that's been a bit concerned for Australians have been the cost of insurance, and I'm rising premiums. And you know, we've seen, you know, the sort of softening of the reinsurance markets seem to be helping, helping that, but these events we've seen the hailstorms end of last year. We've had this wildfire. And how do you think that'll impact the Australian insurance market? We've seen people like some companies like some co-report that they've blown through their reinsurance cover, but it remains a bit concerned, and there were also reports that the group of Queensland Councils have approached Marsh to structure mutual insurer to address the rising insurance rates. So how do you see that play out? Yes, interesting that we were, our attention was drawn to the story initially as the insurance council of Australia emailed us a transcript from their CEO Andrew Hall. He'd given an interview to ABC radio, the national radio on Tuesday this week, and they're the emaileders. Clearly, we've clearly only part of the transcript, because we've only three questions, but they clearly wanted to highlight this issue, it needs spoken quite in depth about it. And basically, he's expressed his disapproval of this proposal, which has came from a group of Queensland Councils. They're trying to create a not-for-profit mutual fund for regional residents and businesses which have been hit by premium increases. Andrew Hall expressed the, he understands the frustration of the local leaders, and both expressed concerns about a localized solution that has no diversified risk and is heavily financially exposed. I think the insurance cost will remain a bit concerned for Australians, and especially as the sort of events that we've seen recently, I mean, apart from the hillstone, we had a big cyclone earlier in the year, we had a flash flood, and that's going to continue to affect people and drive up insurance claims, which would see costs going up. And it's not just Australia, I mean, we have also had a few events in New Zealand or the weekend, especially heavy rainfall in Auckland. Again, the Australian insurers are exposed to New Zealand claims as well. And again, the New Zealand events remind us of the twin events that happened a couple of years ago, the Auckland flooding. I was a big, insured loss event. Do you have the numbers from the New Zealand event, Andrew? So at this stage, they seem relatively small at IG, confirmed they received around 80 claims across home contents motor boat and commercial policies, following heavy rains in the northern region of New Zealand. Further down poll are expected, so that those figures are expected to rise. Interesting in New Zealand's Met service, predicted a month's rain would fall in two to three days in Northland. So that's a lot of rain into one area in short to short space of time. And yeah, we've got some of the ones we're going to have to keep monitoring, because the claims are going to keep rising and the impact of this event is going to continue. Yeah, I think it's going to be a busy summer for insurers down under. Moving on, Russian, you've been thinking into the fast growing data center sector across Asia Pacific. And data centers, as we know, is all that the industry is talking about these days, giving the need for capacity to meet the hyperscale developments we're seeing around the region. The water, the broker, is an insurer telling you? Yeah, like you said, it's been a very popular topic recently. So we're looking at, again, rapid data center growth across the Asia Pacific, which is driven by AI, increased cloud demand and even some massive hyperscale investment. APEC now actually accounts for more than a third of the global data center capacity. And it's only going to continue to accelerate as we go forward. Allian said that the region will see more than a 20% year growth through to 2028 with major buildouts underway in places like Malaysia, Indonesia, India, Vietnam, and Thailand. Singapore still remains a key hub while China and Japan and India actually make up most of the installed capacity. But there are some challenges that are being faced. So there are rising asset values, power and cooling demands. And AI dense compute clusters are creating new high severity risks. Some of these mega campuses also needs billions in cover to push past traditional underwriting limits, especially in crowded markets like Singapore and Tokyo. Intures are struggling somewhat to keep up with that fast challenging technology from the liquid cooling and battery systems. And even leading to which leads to like a tighter underwriting, higher deductibles and more emphasis on strong engineering. To bridge this gap, they've mainly said that they need alternative capitals such as ILS funds or sovereign wealth investors and captives. And parametric solutions are also becoming quite popular with, especially for flood high fumes and grid instability. What's very clear from what I've, the information I received is that the data center boom isn't just reshaping some the digital infrastructure. It's reshaping the insurance market and pushing it to evolve as fast as the technology it protects. Thanks, Orson. Yes, I mean, it was the thing, renewable energy went through the similar cycle, but I think data centers are probably going to be much bigger for the region, the scale at which they're all building out. Thanks, Orson. Joanna, you've been following up on your scoop on Vietnam's new range or a referee, which has just started operations as we understand. You spoke to the team recently. Can you tell us about their strategy, how do they plan to position themselves in such a competitive market and what their differentiator is? Yeah, so this actually followed my previous exclusive report on referee two months ago. So back then, referee declined to give me comments about their launch, but I guess, you know, after receiving necessary regulatory approval, they are more willing now to talk about their company. So I had an official interview with the CEO Lei Huay Nam, and he said that the company is founded by a group of stakeholders in the finance and banking sector in Vietnam, with EVF general finance holding the logistic followed by their first securities and also other investors. So they have a charter capital of 19 million US dollars and they are headquartered in the capital city of Hanoi and currently, Anfari has 15 staff members, including two basic Singapore in order to serve overseas clients. So in the first year of operation, referee will focus on domestic facultative insurance or large scale projects, including renewable energy power plants. This is because, you know, the other two Vietnamese reinsurers, Binari and Hanoi have already deployed their maximum capacity into the market. And at the same time, the percentage of local reinsurance section in Vietnam has been stored at, you know, 12, 13 percent, meanwhile, offshore session increases year and year, and is now at 33%. So this is particularly true for the fact market. And that's why they decided to focus on this segment in order, you know, to help to retain more risks in Vietnam. So, Johnny, this is the third reinsurer's end of Vietnam. Well, what's the reaction being in the market to their launch and their plans? Right. So, since last December, Anfari has started his nationwide pitching campaign to clients and they actually started writing in the last one one renewals. So the CEO told me that the company received lots of welcoming support from clients because, you know, the demand for reinsurance is higher and higher. And clients want to have another source of reinsurance solutions to assist them. And moreover, with the RBC regime gradually coming into effect starting from 2028 in Vietnam, he said that the demand will be even nearer more. So this is a pretty favorable condition for them to enter the market at the moment. It's always exciting to see new reinsurer setup, especially in Asia where reinsurance isn't where the capital first flows to. So it shows you the sort of opportunities in the region, but great to see how Alphari sort of builds itself out in the coming years. And we'll be keeping a close eye on them. Great. So now it's time for us to talk about what's coming up. Aiden, what's keeping you busy? So I mean, so I'm working on a story about the outlook for the catastrophe bond market in 2026. So after a record 2025, the catastrophe bond market, you know, is really in focus at the moment. And, you know, I'm looking into whether the momentum that we saw in 2025 can, you know, can continue into 2026. And based on my conversations so far with various market practitioners, the consensus just seems to be that, you know, this aspect of cost has really grown up. And, you know, the momentum is going to continue to be strong this year. Certainly that's an interesting angle. I mean, I think Caldmont is going to be a big topic coming here. I mean, towards the end of last year, we saw QB issue a new bond with Australian New Zealand Quake events covered. There's also one from an ego where that covers Australia Quake events. So I guess Asia Pacific will see a lot of new issues and certainly what's to keep an eye out is for casualty side cars, which is the latest thing in the block. And I mean, it's probably a while away before any Asian companies would want to do a casualty side car. But I guess it's quite interesting times for the eyeless market in general. Thanks, Aiden. Gerna, what about you? Any big stories in the pipeline? I've always got big stories on the goal. So recently I spoke to a financialised broker and she told me about the rise of parametric products in financialised. So, you know, while these parametric solutions are more established in the property catastrophe space, she's seeing growing interest in data enhanced and quasi parametric concepts within financialised. And this is particularly evident in areas like cyber, VI, linked to technology outages and certain regulatory and operational triggers. So I'll be following up on this to see if other brokers also share the same viewpoints. Certainly, I think we've seen some cyber parametrics and data senders, obviously, a perfect opportunity for parametrics. And I think Roshan's story or sort of mentioned the need for parametric products for the data center sector. Interesting. Great. Roshan, what are you tracking? Keeping in line with parametrics, I've reached out to Celcius Pro, the parametric insurance specialist. I'm chasing their response and hopefully something develops over the near future. Right. And on my end, I'm fairly, I'm at a late start to the new year. I'm still counting up on what's happened at the very old and I'm trying to reach out to some of my contacts. I do have a few interviews lined up with in the coming weeks. I'm really still doing my reporting so I don't have stories that I'm chasing at the moment. There should be something in the work soon. Great. It's been a fantastic meeting, everyone. I think it's a wrap. I know you're all eyeing the clock and waiting for lunch. So I won't keep you any longer. Let's all get back. The guess for this episode of Between the Lines is Dr. Andy Smith, Chief Operating Officer and co-founder of Swiss Rebacked Flood Modeling Specialists Fathom. Andy is the Chief Operating Officer at Fathom as well as an active research scientist with a PhD in Flood Modeling. Andy talks to us about Fathom's technology and how it sort of differentiates itself as a flood modular as well as the emerging risk of flood as a peril and how insurers, governments and flood modellers can help close the penetration gap. Thank you, Andy. Thank you for joining us at Between the Lines. To start off, as a founder of Fathom, can you tell us a little bit about yourself and about Fathom how it all started and what's your mission? Sure. Well, let me start by saying a big thank you guys for having me. It's great to be here. Really, I'm a scientist. That's where it all starts. I did a PhD many years ago now in this thing called Computational Flood Modeling. And the PhD itself was focused on trying to couple flood models with climate models, but really the idea for the company Fathom came from our ties with the insurance market. So at the time, there were many big floods happening actually principally in Asia that kind of woke up the whole global insurance market to the fact that for most of the world, we didn't have coverage. So the insurance market was underwriting risk around the world with very little knowledge actually of what flood risk looked like. So that was really the genesis for the company and led myself and my co-founder Chris Samson to start the company because we were building the first generation of these global flood models. And by global, what we really mean is yes, they cover the whole globe, but the unique thing about them is that these models can be built in very data poor parts of the world. So you don't need local information. So Fathom really attempts to build these models anywhere in the world. And we've gone from models that were actually truthfully pretty crude when we started the company, limited by available data sets, through to the kinds of models that we have now. And the models that we have now are truly a revolution in our ability. Just a few days ago, we launched a new terrain data set around the world called Fathom DM. And in short, that is the world's best image, if you will, of what the earth surface looks like. When we combine that with all of the methods that we have here, Fathom builds these flood models now that are useful for insurers, governments, engineering companies, banks and asset managers, emergency responders. So although we started as I guess an insure tech company, actually we provide information now to a whole range of different organizations. And the company itself was acquired about two years ago by Swissry, one of the world's biggest reinsurance companies. And that really for us is another way to get our data into even more organization's hands, because I really do feel like these models have the ability to really improve societal resilience to things like climate change. Great. Thank you. How is Fathom? I mean, you mentioned that it's a global model and it's not reliant on local data sets. How is Fathom's model different from some of the other modelers out there in the world? How do you differentiate yourself? It's a great question. I can focus on some of the technical things, but then I'll also mention what is one of the true USPs of what we do. So I would say the physics of our models are pretty unique. So how much of the computational algorithms are focused on representing the flow of water for example? So the raw physics is unique. The input data sets that we use and our ability to process them are, I would argue, a massive USP. So I've mentioned just now things like terrain data sets. So we have the world's best global terrain data set, and depending on models, with the only organization, believe it or not, that represents river channels for the whole planet. So there are many model components that are unique. But actually in terms of a company ethos, the thing that really I would say differentiates us is that we are the scientific leaders in this space. And when I say we're the scientific leaders, what I mean by that is that our models and our methods and our input data sets are peer reviewed. So we publish this stuff in the world's best scientific journal. So I can also say that we are the science leaders in this space. That's important because not only should it end dear faith in the methods that we apply, but hopefully it provides transparency as well around kind of how to appropriately use these kinds of data sets. That I think is also very important as we enter the world of non-stationarity. That's another way of saying climate change. Like how do you end users believe what we say with regards to the future, right? You don't have a time machine. So how do you how do you have trust in organizations telling you about what risk is going to look like in 30 years? Well, I think the best way to do that is being completely transparent and letting our peers around the world validate the ways in which we build these models. So that is another, I think, and probably the core USP of our organization. I mean, often, often when we speak to the industry, especially insurers and reinsurers, I mean, there's no doubt that we've seen a lot of flood events and this flood as a peril is certainly becoming a bigger threat, bigger and bigger threat every year. I think we've seen that last year and the year before significant flood events, especially in Asia-Pacific. Obviously, one of the things that people say about, you know, in Asia, especially Southeast Asian, tell the issue that we've seen a lot of flood events, insurers talk about lack of models, lack of data and ability to sort of preemble some of these events. How do you see that challenge, how do you see solving that challenge in the region, especially given the sort of protection gap we have in these regions and the vulnerability to climate change we're seeing in some of these regions in Asia? I think better models has to be a part of the solution, right? Better models enables insurers and reinsurers to better price risk, right? Which has to be a part of the solution. In the absence of good models, insurers will either not provide coverage or they'll be very cautious, so it should be an insurance enabler. Also, making organizations outside of insurers aware of the risks also has to be a big part of the solution, right? You need to make people aware of the risk they're exposed to. And that goes for the general public. It also goes for things like governmental organizations. Like, can we get some of these data in the hands of governmental organizations who in the past haven't had access to them? And so in a kind of an environment with a lack of knowledge that kind of they're not empowered to take action, right? So I think what we're doing here at Fathom is clearly not the whole solution, but I think it's a big part of the solution. Let's get these models that now exist. They're brand new, right? So I often say you, it's, you shouldn't overly criticize organizations for not being aware or taking action around these problems right now because the science is so new, these models are so new, they're kind of, they really are kind of for one to the better phrase, they're fresh off the conveyor belt, right? They're just being built. Nevertheless, they are now here. We have them. They do provide useful information. So let's get them in the hands of organizations and make society, particularly in Southeast Asia, more resilient. I mean, Southeast Asia, truthfully, if we look at our models of global scales and we look at things like our economic loss models, so the models that give us estimates of economic damage, actual damages experienced by floods, Southeast Asia is the most exposed part of the world. You have a lot of assets, a lot of, a lot of urbanization, frankly, in harm's way. It's also a very, again, for once, a better phrase, a very wet part of the world, right? You have things like the Mechong Delta, right? A very flat wet part of the world with lots of infrastructure in the way. So let's get these data in the hands of people and make that part of the world more resilient. How do you see, I mean, we've had some really interesting event. I mean, for the one for another word, but really unprecedented sort of events around around the world, and certainly so in Asia-Pacific. I mean, there was a, I imagine, a flash flood last week in Australia, which was, you know, so nothing that, I don't think that you could have modeled for it. They're the sort of events that we saw towards the end of the year in Thailand and in Indonesia and Sri Lanka where we had those, you know, the senior and DITWA and the sort of all these things coming together in ways that models, probably how, you know, there are no historic models to sort of predict that. How do you see, how do you see finding, you find modeling these kind of new events and unprecedented events and how do you see the impact of that on the broader industry? Yeah, it's difficult, right? Because we are, I mean, escalating climate volatility is clearly happening, right? And actually, the kind of, if you want to break it down to the physics of that, it's actually pretty straightforward. A warmer world, a warmer atmosphere, holds more water, right? This is very old physics, physical law called the Cloutier's Clipper and law, and it dictates that, right? There's hotter air, holds more water, so it will be a more volatile system. So we have a more volatile climate system. We also have models that, I guess, traditionally lots of them have been based on observations, and that is a big problem, right? Because if you're building models based on observations of a system that's fundamentally changed, then it's no surprise that your predictions will be pretty hopeless, right? Because it's a new kind of system. And into that, actually, I would say that even if the system was stationary, so if we kind of pretend that the climate is entirely stationary, the atmosphere itself is a, in mathematical terms, it's a nonlinear system. And a nonlinear system, it gives you chaos if you let it run long enough, right? It's kind of, it's the same reason why weather prediction is so difficult, right? If you move five, ten days out, it gets really hard, because the system is very, very volatile. My point here is that we've not been observing the climate system for very long, only a matter of kind of a few decades. And if you're talking about things like a one in a hundred-year event, well, we've not been observing the system for a hundred years. So even if the system was stationary, using models based purely on observations would be a problem. So in answer to that, again, we need these models that have different ways of representing that uncertainty, right? We need different ways of representing climate uncertainty. We need different ways of filling in the gaps with our observations as well. That's exactly what we do at Fathom. Actually, I would also say that it's where the rest of the world can learn from insurance, because broadly speaking, this is what catastrophe models used by insurers and indeed re-insurers are designed to do, right? They're designed to give us an understanding of these big tail events. So when we're hearing about these unprecedented events that have never been seen before, well, that's exactly the kinds of events that generally exist. And yeah, we do get caught out as an industry. But those kinds of events are generally the kinds of events that are contained within the events that's used by insurance companies to try and understand tail risk. So I do think these tools, yeah, originally designed to understand tail risk in the insurance market. Actually, I think society at Hull can use them and use them to understand some of these new risks that we're seeing. Other ways in modeling to sort of react to the climate change. I mean, we said that it's been based on observation and we haven't been observing for a long time. How do you account for these variables that climate change brings into your modeling process? Another great question. The principle way to represent climate change explicitly is to use climate models, right? So you have models of the climate system. And models of the climate system attempt to simulate what could happen to the Earth's climate over time, right? And there are many of them and there needs to be many of them because we need large ensembles given the uncertainties. But those models attempt to simulate, well, what does the climate look like if we perturb it in certain ways? And those perturbations generally take the form of inputting things like CO2, right? So you ramp up emissions and then you see what the climate system looks like. We can plug those models into our risk models, right? And our risk models will be things like the models of river systems around the world or the models of flash flood systems around the world. And then we can try to close that gap. Closing that gap is very difficult because one thing to emphasize is that climate models, they really don't represent the things that we care about as risk modelers because they operate on entirely different scales, different scales in space, different scales in time, right? So they don't represent kind of risky phenomenon or weather phenomenon as a better way of putting it. Nevertheless, we do do that. So we link those things and fathom provides climate risk information for actually for any climate scenario that you want with uncertainty. The other thing, however, that we do in our models principally in things like our catastrophe models is that we attempt to try and simulate the stochastic kind of uncertainty of the atmosphere anyway. So we try to simulate things like, well, we don't have many observations. But what we do is we try to generate a synthetic 10,000 year event set. So 10,000 years of plausible events, things that could occur. And when you do that, actually, the range of events that occur around the world is pretty vast, right? So yeah, if you only looked at observations from the past 40 years, there's not that many really extreme events that occur, right? If you have a 10,000 year event set, regardless of climate change, actually, that range of events can generate some pretty huge losses around the world. So that's kind of two different ways of representing these things. Insure as often have relied on the stochastic understanding or a synthetic event set to understand what tail risk looks like. Often now, what we see from regulators is they actually want you to explicitly explore emission scenario risk, right? So what does my risk look like if we were to be in an RCP 8.5 world in 2050, for example, our tools do both of those things is really what I'm trying to say. I mean, we've spoken about flood and flood as apparel seems to be growing threat around the region. I mean, seeing more and more flood events and, you know, both flash floods and rubberine floods and all kinds, you know, flash floods as apparel in itself is growing into a bigger risk. And sort of insurance penetration, as we know, as we always hear from insurers in the region is it continues to be very low and that sort of protection gap continues to widen in the region. How do you see how can sort of better flood data and modeling help us close that flood protection gap and how can governments insurers and technology players like Fathom solve for that problem? I am definitely in danger of repeating myself. So I will, I'll apologise and ask you to forgive me, but I do think it comes down to this, the requirement for better data and better models, right? That makes these risks more. It allows us to represent them in more accurate ways. We try and be transparent with how we build these models. But in that and in the environment in which these data exist, we need governments to play their role, right? We need governments to help close up protection gap. They should be improving how these data are shared, right? So the public are aware of the risks that exist. That in and of itself should drive demand. We also need governments in this new world where we have climate volatility and we have rapid urbanization that in fact is at least as kind of amplifies risk by at least as much as climate change and truth if not more is what the evidence shows us. In that in that world, we need governments to share data. We need them to invest in things like defenses and adaptation and we need them to support things like public private insurance schemes for really high risk areas, areas in which without that it's very difficult to provide insurance, right? So we need that. We also need insurers to have access to these models as well. For the reasons that I mentioned previously, we need insurers to have good information with which to price risk. Because in an environment where they can price risk more accurately, they don't need to be conservative, right? The insurers need some way to effectively and accurately measure the risk with which they are exposed to. And that is really again, and I apologize for repeating myself, but that's really where I think technology providers like Fathom has a big role to play because we're the ones generating the information to do all of those tasks, right? And I think we really do without sounding too grandiose. I think we have a real role to play here in making society at all as a whole more resilient to these new threats. Just from a Fathom perspective, you focused on floods that all you focused on, did you start looking at other perils around the world? Well, generally we're focused on flood. Really, that's our core, again, USP. I mean, early in the early days of the company, if I'm being truthful, yeah, we thought about it, right? Because people said, well, you do flood, why do you need the other things? But we kind of decided early on that. Now, we're just going to, like, this problem is so big and difficult to tackle that it requires absolutely all of our attention. And what we'll do is instead of trying to provide an answer for everything, we'll try and provide a very good answer to this one very specific and growing problem. So we have around 60 people now, only building flood models. So that's what we do. Nevertheless, some of the data that we produce to improve our flood models are useful beyond just flood, right? I mentioned the terrain data sets. So we have the world's best terrain data sets. Now they can be used by other modelers and they are indeed used by other modelers. We have explored and done work in areas like drought. So it's not too much of a surprise to say that the models set up to simulate flood processes are indeed useful for simulating low-flow environments as well. So they've been used for things like mapping, shipping route risk around the world. So what are the chances, for example, that you're not able to ship grain to the Midwest or the US or through South America? So they are used for those things as well. But as a whole and in general, we focus on building better flood models. I'm really excited and actually amazed to truth that I'm amazed by how it progressed we've made from the early days of the company to now, because we have locally accurate flood models anywhere in the world. So we have made huge strides, but there's still a lot of work to do as well. So I don't think we should get complacent in that. We still have much work to do and there will be even more exciting news in the near future. Thank you, Andy. Thank you very much for your time and for your responses. It's been fantastic speaking to you. Brilliant. And again, thank you so much for the time. That's all from us in this episode of Between the Lights. We'll be back again in two weeks. In the meantime, you can connect with us on LinkedIn and make sure you're subscribed to our daily and weekly newsletters, delivered straight to your inbox. And of course, we'd love to hear your feedback on the podcast. Email us at editorialadinterinzationnews.com. I'm Mittan Markey. Thanks for listening.