In this episode, Charlie Samolczyk is joined by Chris Bird and Charles Carwin to discuss the trends and themes they have seen in the P&C capital modelling market.
CHRIS BIRD: I always think that capital model's a bad name for it, because it's principally used for capital assessment. But it's so much more than that. I prefer to call it a risk model, because what it is really, it's a holistic model of the risk of the business.
SPEAKER: You're listening to (Re)thinking Insurance, a podcast series from Willis Towers Watson, where we discuss the issues facing P&C, life, and composite insurers around the globe, as well as exploring the latest tools, techniques, and innovations that will help you to rethink insurance.
CHARLIE SAMOLCZYK: Hello, and welcome to Talking Technology. I'm your host, Charlie Samolczyk. And in Talking Technology, we explore the wide range of technology issues facing insurers. From AI and data science, through to open source solutions and cybersecurity. And we look at how we are helping our clients to tackle these issues.
Welcome to another episode of Talking Technology, a WTW podcast. Today, we will be covering PC capital modeling and some of the themes and trends that's going on in the industry. I am thrilled to be joined by Chris Bird, who is our global leader for the PC capital modeling proposition and practice, as well as Charlie Carwin, who is our Americas proposition leader for the PC capital modeling. So welcome to you both.
CHRIS BIRD: Hi.
CHARLIE SAMOLCZYK: I always kick these off with a little bit of a random question just to get things going. So I know I haven't given you any pre-warning on this, but let's talk about superheroes a little bit right now. My son's been watching some Spider-Man gaming stuff. So just to kick things off, you know what's coming. If if you had a superhero power, what would that be? And maybe as a part B, would you apply it to your insurance practice? Probably not, but maybe with that. Chris, I'll start with you.
CHRIS BIRD: I always feel like-- live forever. Maybe not live forever, but I'd like to see in the future. I'll be fascinated to see what's going to happen to the world in 200 years' time.
CHARLIE SAMOLCZYK: Not so different from mine. I think mine, I would love to be able to stop time. So I don't know if that's the same thing. But it's a little bit of getting more out of the time that we have or seeing I guess-- maybe it's a little bit different.
CHRIS BIRD: I guess I just think with things like climate change and artificial intelligence and all these developments, and some of them are quite scary.
CHARLIE SAMOLCZYK: Yeah.
CHRIS BIRD: And it may turn out to be a disaster. But actually, it may turn out that the human race does overcome it, and I'd be fascinated to see that. So I don't know, maybe not specifically a superpower. But I'd be, yeah, fascinated to see.
CHARLIE SAMOLCZYK: There you go. You can also put time travel into that as well.
CHRIS BIRD: Time travel, there you go. Superpower, time travel. Charlie, over to you.
CHARLIE CARWIN: That's a fun question. I think I'm going to try to keep it smaller, because I think time travel, I'll probably get into trouble somewhere. I would just like to be able to eat whatever I want with no negative health consequences. I think I'd be pretty happy just being able to have unlimited donuts and pizza and so on and feel great for the rest of the day. So that'd be fun.
CHARLIE SAMOLCZYK: Yeah. Very cool. OK, well, let's switch gears and bring this back to PC capital modeling. So I mean, you guys are experts in this space. Work with a lot of our clients, have a hand in our technology and what we're doing from a proposition point of view. Capital modeling, in my understanding anyway, has changed over recent years in approaches and as technology has ramped up and is supporting, I guess, the practice.
Based on your experience, though, just to set the scene, what are some of the key challenges in your mind that the industry is facing right now around capital modeling?
CHRIS BIRD: So I guess capital modeling, if we look back over the last 20 years of how it's developed, in the early days, there was lots of developments on methodology and techniques. And I think, and I think Charlie agrees, we've discussed this quite a lot before, that people perhaps got a bit carried away and a bit over complex and built big monsters and things. I mean, that created a challenge perhaps 10 years ago of usability and to be able to use models well.
So I still think that's the real focus today, really, is just trying to make models more usable, get more value out of them. It's less about making them more sophisticated. In fact, people have said models have probably got more simple, really, in recent years. But the real focus has been on usability. Doing more with it, doing it quicker, responsiveness. Not having to wait hours or days to get an answer to a simple question.
CHARLIE CARWIN: I think I'd pile on with that, Chris. And I definitely agree. I think capital modeling is probably easier to do than it has at any point in the past. So the tools are getting better. People understand the methodologies. It's not necessarily a new science anymore. So I think that makes it a lot easier to do things. But I think as an industry, we still struggle to extract value or the amount of value that should be extracted.
I mean, I always say that the capital model is just this amazing piece of technology. That it encompasses all the best assumptions from around the company. And nowhere else does that exist. You have all the different components coming together in a central repository. And you can just see what happens with the company with simulations. It's fantastic.
But I question if the industry does get enough value out of that for all the capability that it can offer. And I think technology is definitely helping that. And I think that's going to help address the questions, and I think that's one of the reasons I'm excited to have this discussion.
CHARLIE SAMOLCZYK: And maybe I should have asked this out of the gate. But maybe really succinctly, for those people listening that aren't capital modeling practitioners, what is the function of capital modeling, and what value does it bring to the organization?
CHRIS BIRD: As Charlie says, it's perhaps the best central model of the whole business. It captures everything. It captures the asset sides of the balance sheet as well as the liability side. So a capital model, I always think a capital model is a bad name for it, because it's principally used for capital assessment. But it's so much more than that. I prefer to call it a risk model. Because what it is really, it's a holistic model of the risk of the business.
And so to be able to understand the risk of the business, well that's useful for various things. Principally, how much capital do you need to run your business? But also for management to understand their day-to-day volatility. We're seeing a lot of that at the moment. Reinsurance prices have gone up, people are raising retentions because they can't afford to buy as much reinsurance. But therefore, are they taking on more volatility than they had previously?
Well, again, the capital model is the tool which you can use to help understand your volatility and how different strategies or business levers affect the risk of your business.
CHARLIE CARWIN: Yeah. When I describe it to others not in the insurance industry, I say that we're building a model. And already, I think lose a lot of them. But essentially, we're building a tool that can answer what if questions for what if happens to the insurance-- what happens to the insurance company if something happens? And I think folks even inside the industry take a look at the capital model in their mind and they say, wow, this is a really complicated thing.
And it can be. But at the end of the day, all the calculations could be done on the back of a very large napkin. It's things flowing one direction, how it impacts the financial statements, how different things are simulated and so on. So it's this tool that says, what happens to insurance company if X, Y, or Z happens? And then let's do that over and over and over again to see where our risks are for maybe a different lens. So it's really just a fantastic piece of technology that's been developed.
CHARLIE SAMOLCZYK: And Chris, I know you've presented on this in lots of different forums. But trying to, I guess, maybe move away from the silo of just the capital team using the model. Maybe do you want to elaborate on that? Where do you think the capital model needs to flourish into different parts of the business?
CHRIS BIRD: Yeah, it comes back to that thing on usability. Until recently perhaps, the capital model has really been the preserve of the capital team itself. A team of specialists who know the model and can operate it. And that's fine. Capital people, capital teams, they're good people. They try and be helpful to the business. But what it means is that if the reinsurance manager wants to test the reinsurance, they have to ask the capital team to run it through. So they have to rely on the capital team being there.
Maybe it's reinsurance renewal time just before New Year. The capital people might be off on their Christmas holiday. So it's not always convenient to have to rely on the team, have that team of experts there to run things through. Or you think of an underwriting manager who wants to test out different business mixes and portfolios and things. And they may want to just be testing this over and over again, all sorts of scenarios and things.
And currently, they'd have to go to the capital team, give them the information, and get the team to run things through. And it's fine. Like I said, the capital team, they're generally helpful. But you just think it would be so much better if these people could be running it themselves. We're seeing a bit of a trend of this change in technology is enabling us to create better portals for these users into the system themselves.
So it's not as if they need to understand or run the full model. Every last bit of it or the complex copulas and things like that. A reinsurance manager, they just want to be able to take something that's already set up, put a reinsurance program on top of it, run it through and see the output for that specifically. Not the not the output for everything. And so yeah, we're seeing this kind of operation, well, become more of a thing.
Like I said, you have almost a safe space where that reinsurance manager can interact with the tool directly, run their piece. But they're not burdened by having to understand the full system. So that's a bit of a trend which we see is just really bringing better socialized in the model I guess, or just bringing more users more directly in contact with the model. We're also seeing better connectivity as well between systems. So it's not just the reserving team sends their data to the capital team who runs it through, or the finance team sends their data to the capital team.
Better connectivity between the systems so that finance machine can talk directly to the capital machine and pass the information back and forth between them. Some really interesting developments there to bring usability to the fore. I think it all builds on perhaps recent trends. You go back over the last five, 10 years, there's been a lot of technology developments with cloud computing in particular, which has meant that models have got so much quicker. We can scale models to run over a lot of compute power. You're not just limited by the laptop in front of you.
So models now run really, really quickly, principally because of what cloud technology has done. We've also had a lot of developments in automation. So you're not slowed down by humans checking data, crunching things, and making errors. So both automation and cloud scalability has meant that just the capital model as a system can be used and run through far quicker than it used to be. The trend we're seeing now, then, is bringing in more users to use that. It's not just the capital people pressing the buttons of the machine. You can get these other business users and other systems also working directly with it.
CHARLIE SAMOLCZYK: Do you think that-- I mean, you talked a lot about technology there, which was great. This is the Talking Tech podcast. Do you think that the industry is now waking up to the intrinsic power that's there? Now that we have the compute, now that the automation is there, now that the usability is there, is the industry shifting and saying, hey, actually? And I think, Charlie, you mentioned this. There's this model here that actually we can really use across the business, right, to run some really interesting what if scenarios. Or if situation x happens, how is that going to affect our balance sheet?
CHARLIE CARWIN: I think it's been trending in that direction. I think we have a little ways to go. As Chris mentioned, we talk about this a lot, and we just had a couple of presentations and a blog post about the socialization of capital model. And to put it succinctly, it's other folks in the organization understanding and buying in to the analytics produced by the capital model. And that's a pretty big ask to do.
And everything that Chris mentioned as far as having other folks in the organization be able to get hands on to the capital model, that's a new way of approaching socialization. As Chris mentioned, the reinsurance team being able to see the assumptions and alter the assumptions and do their own what if type analysis, it naturally has a buy-in to the capital model. Same thing with the asset model team, same thing with a lot of different experts around the business.
So I do think it's trending in that direction. And it's all powered by technology. A lot of this stuff that we were talking about right now wouldn't be possible in the old ways of doing capital modeling or something that's not cloud enabled. So they go hand in hand.
CHRIS BIRD: I think as you said in your question, there's an awareness thing. I think if we look regionally around the world. So capital modeling has been big in Europe in particular, because of a regulatory drive. The capital model is what companies can use to calculate their regulatory capital. So these European firms who've taken advantage of this, they're very fluent, I guess, in the use of their models and are aware of the benefits and what they can do with them.
Now, it's interesting you compared to the US. There isn't that same regulatory drive. But certainly the big firms there all saw the benefits of capital models for understanding risk profiles and risk appetite and understanding control and volatility and these kind of things. So these firms have always understood it. But I think we're seeing now, yeah, a wider appreciation. In Europe in particular to perhaps smaller firms who were aware of this, but always thought implementing a capital model was too big, too much effort. The barriers to entry were too high.
When, again, now with technology, it's become a lot easier. So we're seeing smaller European firms taking it on and seeing that they can get the benefits from it. And similarly in the US as well. The smaller firms who might previously have thought we need to have a team of experts, well they now realize that they can operate these things, these systems and get benefits of it without having to have a big specialist team. Just because they're just, again, the usability and easier to operate.
CHARLIE SAMOLCZYK: So in your minds, there's a bit of a shift happening. So why WTW? Why should clients who are thinking about either the-- I don't know if transformation is too big a word, but the way that they're transforming, how they use the model or think about the model, or how they're getting business benefit out of it. What can we offer as a company to help partner with insurers to get the value out of that part of the business?
CHRIS BIRD: So well, just a why WTW. I mean, the principal reason I joined WTW back in 2019 is I really saw WTW certainly in this space as absolute thought leaders, innovators. The ones who are really leading the way in developing and enhancing these things. So yeah. All these technology developments that I've been talking about, it's what we live and breathe.
And it's great. I mean, I had the team that develops our software for this. And it's just, it's fascinating. It's great seeing it developed, seeing it come into life. Discussing these ideas and these things with clients and then taking it back to the team and building it and making it happen. These things are real. I said about the cloud technology grids and things like that, being able to speed things up. And that was something we brought in a few years ago now.
Our clients are using that, and they've benefited massively from it. We won an award last year recognizing that. We have clients who run-- previously before cloud technology, you would run a model on a machine. And it would take however long it would take. But with the cloud, you can run multiple machines. So you can run multiple models all at the same time. So instead of if a business wanted to test out five different what ifs, and previously you'd have to queue them up, run them up one after another. Now you can run them all at the same time.
So yeah, five things may not take too long to run one after another. But we've had people who've run hundred different alternative strategies at the same time. It's rapid compared to, well, previous times when you might have to queued it up and run it over a weekend.
CHARLIE CARWIN: Yeah, I'll just pile on to that. I think the level of innovation that we're seeing at WTW is just, it's really encouraging. Chris just mentioned the hundred different scenarios. I'm not sure that that's something we were thinking about a few years ago. But what we did think about is we know that we need to move to the cloud, we know we need to have scalability. And then we're going to see where the extra features and the extra capabilities takes us.
So in a lot of ways, we're scratching the surface on what is capable and how the capital model can be used to inform different decisions as the technology keeps getting better and better it starts opening up new avenues that maybe hadn't been previously considered but are going to be the way that folks approach capital modeling in the near future.
CHARLIE SAMOLCZYK: Very good. So if I can summarize, Charlie, maybe the capital model's like the all you can eat model, and Chris, you use it to see into the future, just to link it back to your superhero-- no, I'm joking. One of the things we haven't talked about, and it does seem to come up in a lot of the other podcasts that I do, is the notion of open source. And the impact that that's having on different areas of the business. Is that a factor in the capital modeling space? Or how is that influencing how you're approaching work?
CHRIS BIRD: Yeah, so we do see a lot of people when they start out, they build up models. I mean, you can build simple models in Excel, actually. And you find that Excel quickly isn't powerful enough. So then people move over to Python or R, which people come out of uni and they're very familiar with it. What you do see, though, is when you want to get to a proper system, proper industrial system, I guess, or a professional quality assurance system, you need something more.
Although there's plenty of people out there who can code in Python, actually there's a few out there who can code in Python in these detailed, stochastic simulation models, Monte Carlo models, with all the memory efficiency of that. So it's not as easy as people may think. But yeah, you can build up. You can build up models. And they're great for testing and sandboxes and things like that.
But then the key thing is, I guess, once you want to get to a proper professional system the business can use. Like I said, that a reinsurance manager can run or an underwriting manager can run. You need a real professional system. So you need the governance around it. The quality assurance. You need a user interface. You need the version control. So you don't get lost in different sets of data, and where did it come from and being able to not being able to reconcile numbers and all that.
So yeah, I see open source of these. These open platforms, there's always been a great playpen or sandbox and things. But for when you want a real system, they just don't give that. You don't have all the infrastructure. You don't have an interface or a portal in the same way as properly designed, dedicated software does. Our head of engineering in the team gives a crazy stat about the number of hours or person hours of development time that has gone into our software.
And that really does count for something. Particularly all the quality assurance and testing that happens that goes on. So lots of effort to build professional tools, really.
CHARLIE CARWIN: I'll go back to my example of the back of a napkin. Because while the calculations themselves should be straightforward and they're hopefully understandable to folks involved, the back end associated with making that efficient and user friendly, is not simple at all. I think any sort of software development, especially when you have the user experience in mind, the bar gets raised very, very high.
I've been really interested to see with both automation and working with capital modeling teams and larger companies really how much governance comes into play. And I don't think governance is something that can be implemented easily. Something certainly that's very robust and will pass the standard. So there's a lot more than just the back of a napkin calculations.
And one of the nice parts is that with a commercial product like ours, then we're doing all that work in addition to financial statements, in addition to saying, hey, one particular company is looking at X, Y, and Z. Let's go ahead and build that so everybody can look at X, Y, and Z. So there's a lot to be offered.
CHARLIE SAMOLCZYK: And Chris, maybe going back to your superpower of time travel and moving into the future to see how the world's turned out. Maybe not 200 years. But if you went ahead 30 years, where do you hope this practice will have evolved to?
CHRIS BIRD: 30 years is-- OK, it's quite hard to look out 30 years. But do have fairly clear visions of where things are going in the next five, 10 years. It's interesting seeing tools such as generative AI and more generally AI coming in. And we believe this is going to give more power to go further with things.
As I said, we've seen-- if you look back over the last 10 years, models haven't got more sophisticated. Actually, I think they got simpler and become more usable. I think the trend we're going to see over the next five years is granularity. So typically people run models maybe a line of business level, maybe a product level. Some reinsurers may be run down at contract level. But I think more and more, we're going to see that kind of granularity becoming more widespread through commercial insurance, insurance more widely.
OK, I don't think we'll have personal lines insurance running every single individual person and their car and their things like that. But I think, yeah, we're going to get a lot more granular right down to contract level, certainly for commercial lines or product level for personal lines. There's a real demand for that. I mean, you think about key aim for a business, it's return on equity, return on capital.
And so when firms are pricing, how do they do their pricing? What profit margins should they be targeting? And ultimately, if you think about it, it comes back to you're trying to set your pricing to roll up to make an overall enterprise return on equity. So actually for pricing, what you really want is you want to know how each contract contributes to the capital needs of the company. We get this question asked all the time of, well, how can we allocate capital down to contracts?
And that's hard to do. Our models generally aren't granular enough. But we're seeing things going that way. Technology is giving us the power to go that way. And we just need to work a bit more on the data, I think, and some of the techniques to bring the modeling on as well. But yeah, that's a key trend I think we're seeing over the next five or so years. So far more granular outputs from modeling.
CHARLIE SAMOLCZYK: And maybe totally a question to you. So maybe a final question. I think for those people listening that are a little bit on the fence, maybe they're still working in Excel, or to use your example, they've got a very large napkin in front of them where they've been doing this. What's the urgency? Why act now? Chris, you talked about some of the business benefit. But what are you hearing from clients as to why they're taking a serious look at this now?
CHARLIE CARWIN: So I'm always encouraged when clients or companies, whether they're on an Excel model or a competing product from WTW or whatever they're using, I'm always encouraged when they're using a capital model and they're getting value out of it. So I think that's the most important step, especially in the Americas where we don't necessarily have the regulatory framework that requires it. I'm really encouraged that folks are looking at their risk in that way. They're asking those what if questions. They're having those discussions.
If they are on technology that's maybe a little bit more homegrown or a bit more difficult to maintain, in large, I would guess they would actually have some cost savings and some time savings going for a robust commercial product that has had decades of experience poured into it and how to make it easy to use. How to reflect different risks and so on. So capital modeling, especially from a commercial standpoint, I think the barrier to entry is much lower.
And that's intentional. We've taken this very complicated system and we simplified it such that the size of the team is very small required to support. The technology is extremely efficient. So I think anybody doing capital modeling right now is on the right path. And I would just encourage them to question whether or not there is a more efficient way to do what they're doing.
CHRIS BIRD: I'll just jump in to say. I mean, hey, the competitive landscape out there, there's key benefits you get from these things. In Europe, you can see real capital savings. That's a competitive advantage. OK, in US, you're not going to get regulatory capital savings. But controlling or understanding your portfolio risk and managing your portfolio in this way will lead to rating agency capital savings.
Reinsurance has got expensive. How do you know what is the best program for you? Again, it's a competitive thing. People who can properly analyze their reinsurance needs will most effectively deploy their spend. They will outcompete those who don't. So I just think there's some real clear competitive advantages from doing this.
CHARLIE SAMOLCZYK: I feel like reinsurance optimization could be its whole own podcast. Excellent. Well, I think we're out of time. But I would like to thank the both of you. I think it's been a really interesting and engaging discussion. And hope to have you back on another podcast in the future. Thanks very much.
CHRIS BIRD: Thanks, Charlie.
CHARLIE CARWIN: Thanks, Charlie.
SPEAKER: Thank you for joining us for this WTW podcast featuring the latest perspectives on the intersection of people, capital, and risk. For more information, visit the Insights section of wtwco.com. This podcast is for general discussion and/or information only, is not intended to be relied upon, and action based on or in connection with anything contained herein should not be taken without first obtaining specific advice from a suitably qualified professional.
For over a decade, Charlie has provided transformational insurance solutions with specific focus on automation, cost reduction, internal efficiency, system integration, legacy system modernisation and distribution channel connectivity. Charlie is acutely focussed on ensuring our solutions and services exceed client expectations.
Chris joined WTW in 2019 and is the Global Lead for P&C Capital Modelling. He is responsible for leading WTW’s P&C capital modelling proposition, with a focus on the software product, Igloo. He is recognized as a thought leader in the Institute and Faculty of Actuaries (IFOA), and is a frequent speaker at industry events.
Charles is the P&C Americas Practice leader and leads the Americas P&C capital modeling proposition. As part of his capital role, Charles works with clients to help quantify answers to the many what-if questions that go hand-in-hand with robust risk management teams.