The Soft Side of the Soft Cycle | Tony Buckle & John Carolin, UWX

by Juan de Castro, Cytora COO

In this episode of Making Risk Flow, host Juan de Castro is joined by both Tony Buckle, former Chief Underwriting Officer of Allianz Global Corporate, and John Carolin, former CEO of B3i. Tony and John now work for UWX, a company set about reimagining underwriting.During their discussion, the trio talk through effective ways to navigate a soft market in the insurance industry. They highlight that, during the ‘fog of war’, a soft market is not something that just happens, rather, it is something that you can actively manage and be successful in. Whether that be from what you choose to write, to how you choose to manage an organisation, and how measures you use to quantify both.

Listen to the full episode here

Juan de Castro: Today, I'm really delighted to be joined by Tony and John from UWX. Today, the episode is going to be very much focused on all things around a soft market. Both Tony and John are experts in this space. So let's just jump straight in and, perhaps starting with you, John, if you can give us an overview of your background and the work you're currently doing at UWX, please.

John Carolin: Thanks, Juan. So, by way of professional background, I'm a chartered accountant and chartered financial analyst with a deep passion for technology. I started working in the insurance and reinsurance industry about 21 years ago with PricewaterhouseCoopers in their insurance cluster. I later worked for Swiss Re, who were one of my main clients while I was at PwC. I've taken tours into other industries. I've stepped out of insurance. I have started technology ventures in media, in real estate, and property portfolio management. And then about seven, eight years ago, I came back into the industry. In my most recent role before UWX, I was the CEO of B3i, the insurance industry blockchain consortium owned by a number of large carriers. At UWX, we're one of the founding partners, together with Tony. We focus very much on commercial insurance underwriting. Chief underwriting officer is our North Star, we focus on portfolio management, technological change, how insurtech and third party data can help in commercial insurance underwriters. We've been doing a large body of research into understanding the motivations of underwriters during the soft market, which we'll go into a bit more detail. And I think Tony, although he probably doesn't need much of an introduction, probably can tell you a bit more about himself better than I can do that.

Tony Buckle: Yeah. So I'm Tony Buckle. I've wanted to see you again, Juan. So my background is similar, steeped in insurance and reinsurance. I joined the industry the year before the terrible events of 9/11, when the market was struggling actually out of its previous soft market. So the late ‘90s was a very soft phase in the insurance and reinsurance market. And then worked with GE Frankona during a period before its acquisition by Swiss Re. I then joined Swiss Re and was the owner of leading their engineering construction practice for a few years, then moved to the corporate solutions side of Swiss Re and ran their EMEA function before joining RSA and helping them navigate the aftermath of what had been a pretty poor period of performance in the soft market before joining Allianz and helping them move out again of that soft market phase. So, a lot of experience of soft market portfolio navigation, re-underwriting portfolios into soft markets. So hopefully we can talk somewhat usefully about learnings from that.

Juan de Castro: Yeah, definitely. It sounds like you've got almost a perfect background to get deep into all the considerations around the soft market and what it means for the industry. So let's start. You touched on this piece of research you've done recently, which contains really, really interesting insights. So perhaps, Tony, if we start with an overview of what was the motivation behind that piece of research, and what are some of the key takeaways from it, and what's been the reaction from the market?

Tony Buckle: To be honest, it started just as an area of deep personal and professional interest, to be honest. I mean, I've seen organisations, good organisations, come unstuck in soft market conditions and wanted to, I suppose, put a little bit of the academic hat on again and understand why. And I teamed up with Swiss Business School, which is one of the business schools here in Switzerland. And through my sort of network contacts, particularly the engineering construction community, forged, I suppose, an alliance at one level of forces with EMEA. They supported us as we moved forward with the research, EMEA being the engineering construction sort of underwriters, sort of the industry body. And we went through two phases of research. So we went through a sort of questionnaire, which we put out to the broader community. And that had, I think, 120 responses from the community. And then we've done follow-up interviews in the second phase, where we have really drilled down into some of the themes that came out from the survey and then looked at some of the ideas of that very experienced and, shall I say, scarred community had about how they could navigate the soft market better in the future. So it's a broad topic. We haven't tried to funnel it too much. But at the same time, some of the insights have been really interesting and different maybe to what we originally expected.

Juan de Castro: Yeah, so it's very much focused on what we learnt from the previous soft cycles, hopefully to inform us what we can do better this time around, right? And help those organisations navigate.

Tony Buckle: I think that's absolutely right. And I think the interesting angle for us wasn't trying to look at the capital flows in and out of the industry or the deep finance aspects. We were very much focused on listening to the voice of the underwriters and hearing their explanations for what they'd done, why they'd done it. I don't actually have much time for proving causality. I'm much more interested in hearing what people just say about what they've done and why they've done it, because ultimately, that's the reality. So perception is reality. And they've obviously had a pretty tough time at times. And I think we need to understand their world and put ourselves in their shoes rather than try and sort of analyse it super academically from the outside and say that people didn't really act other than because of a capital move or a capital flow here and there. That just doesn't suit the way that most people look at the world.

Juan de Castro: You're touching on more the behavioural side of a soft market, which is often overlooked, right? Often we talk about what is the role of the actuarial discipline and pricing, but often in the end, the culture and behaviours underpin many of the actions that didn't go right in previous soft cycles. There are four or five of them that I thought were particularly interesting, and perhaps we can focus on those and any others that you might want to add later on. But one of the areas you touch on in that research is about price adequacy and how it's almost like a double-edged sword. And you get into more the refinement of what it really means. So let's start there. And if you can share some of the insights from the research in that regard.

Tony Buckle: Sure. When you look at price adequacy, you've always got to think about adequacy in respect of what? It's a relative measure. It's not an absolute measure. The big insight, I think, from our research was that a lot of companies have quite technical views of price adequacy, and they will measure themselves rigorously on their technical view, without realising that the technical view of risk changes over time. So if you're using a pricing tool to determine what the technical price is for a particular exposure, or you're using measures like long-term price adequacy, those measures in themselves are highly influenced by the way you use the tool. So what we saw actually in even the depths of the soft market, the prices remained at 100% price adequate, even though year on year the premiums were obviously decreasing, often double digits. And the reason for that, Juan, was that people needed to sharpen the pencil or whatever other analogy you want to use and work out ways to provide discounts, use the tools, break down risks, in a way that actually gave them the answer that they needed to get out. And this is not a non-technical line of business. Our focus was the engineering and construction community. These are very, very technical underwriters. And even they succumbed to the temptation or maybe the pressure to actually change the way they use the tools over time so that they got the answer that the community wanted. When I say the community, I don't think we should just say, well, it's what the brokers and the clients wanted. There was often a personal angle to that. There was often a management angle to that. And that configuration of forces led to some compromises in terms of the practices being used to use those tools.

Juan de Castro: Let me just dissect that almost to the next level. Two questions combined. The first one is you mentioned the technical price was quite static and you said it should change over time. When you say it should change over time, is it because it should be adapting to new information on how the line of business is evolving from a loss experience? Is that why you're thinking it should evolve? And when you say the underwriters could be at some point gaming the system, that should not be so much the technical rate. The technical rate should be the technical rate. And then it's what rates do you give to the broker, right? So I would like you to comment on those two aspects, if that's all right.

Tony Buckle: Ultimately, let's start with that thing about the technical rates, right? I mean, the technical rate is a product of three things. It's a product of the exposure that you've taken on, it's a product of the wording that you want to provide for that particular exposure and things like, I suppose the client relationship that you have with those particular individuals and the specifics like loss history and that sort of stuff. Now, if you bring those three forces together and you look at them, you can put those elements into a pricing tool and should be able to churn out a technical rate. And that is what is done. Technical rate isn't independent of what the underwriter puts in. So the underwriter then decides, well, this particular client or this particular team at this particular client is very good. I'm going to provide a discount for that. Or this particular situation, this particular exposure, I feel that it's better than its peer group. Therefore, I'm going to discount for that. So you end up making all these adjustments to the tool, which still provides a technical rate. But of course, if you're trying to get to a particular price, the temptation is to find those areas where you can dial down or maybe dial up the discounts, right? So I think it's that behaviour, which is very subjective, of course, is having a big influence on what is seemingly, for most people, an objective measure, which is what the technical price is. And of course, we've got the inverse happening now, in the hard market, that the market will bear a much higher price than is necessarily required by the tool. So people will load up risks to the things that we've just discussed. So you end up in a situation where the tool itself and the output tool are pretty subjective, but the mistake being made is that people are considering that to be an objective measure that doesn't change over time. And of course, it does.

Juan de Castro: It does. So this goes back to the conversation we had just a couple of minutes ago. It's less about whether the pricing models are right or wrong. It's more about the behaviour of how underwriters use it.

John Carolin: I was just going to add the evidence that we have in the conversations we've had, we've heard some pretty sophisticated methods that underwriters know to get the answers that they want. There's this double-edged sword. You build a better pricing tool, more flexibility, more flexibility creates more scope to game the tool if you know how it works. So I think that's a very interesting insight to look at the behaviours around pricing tools, not just the outputs.

Juan de Castro: And as part of your reflections on this research, did you get any recommendations on how to avoid it or how to mitigate that?

Tony Buckle: Yes, I think the most important aspect, Juan, is recognising that fact and not pretending that the technical price is objective and is objective over the whole cycle. I think that's an important and key insight. So if you're looking for objectivity, maybe look elsewhere. There are other measures that we would recommend that people look at for objectivity around where the market is rather than the technical price that's being generated by their own organisation. I think there are also some interesting insights around tool flexibility. I mean tool flexibility itself is a double-edged sword. You want tools to be able to reflect the sophistication of the underwriter, the experience, and the expertise of the underwriter. But of course, flexibility can be used and it can be abused. So you end up in a situation where some organisations are trying to slim down their tools and make them more rigid, I suppose, in terms of the output they can, or maybe better put, in terms of the inputs that they will allow to be varied. So the discounts cannot be applied across multiple dimensions and therefore you get a more consistent price out of the tool. So yes, we've taken a lot of lessons out of this and I think there is definitely scope for best practice in tool usage going forward. And as I said, it's focused as much on the behaviour as it is on the tool.

Juan de Castro: If anything, this also raises the importance of having the right tools, as you just said, I'm just playing this back, as well as having the right culture, incentive, and management structure. We'll talk about management structure, culture, and communication in just a second. One other element that I thought was very interesting in your research is this concept of data-driven versus data-justified. And I would love you to comment a bit more on what you mean by that.

John Carolin: I think we saw sort of two polar opposite responses. There was this one response that we got, which I would characterise as being filled with quite a bit of hubris, which is, we've got better data now, this will never happen again. But by and large, the survey respondents in our interviews revealed that people had the data then. They explicitly knew that the business that they were writing was unprofitable. So it wasn't a lack of data, it wasn't a lack of data quality. They speak about, how to participate in a risk, they needed to find reasons that they could. And the more data they had, the more data points they could find to support why the risk was acceptable. And so I think this goes into culture. How do you make your decisions? Are you looking for more data, more valid, accurate, and complete data? Or are you a data-driven organisation? There were a few respondents who spoke about automated underwriting or algorithmic underwriting as maybe the only way to fix this is if the computers make the decisions and then we review the decisions, rather than people making decisions and finding the data to support them. So I think that's a very important insight that we take away from data and analytics from the work that we've done.

Juan de Castro: Is part of this, so you're just talking about underwriters knowing a risk was not profitable, but finding reasons to still write it. Does it come back again to internal pressure or external pressure? Internal pressure from management to grow at a certain rate or external pressure from the broker almost asking for a favour?

John Carolin: It's a number of factors, and you've touched on some of them. We looked across a broad range of factors, client and broker pressure, management pressure, incentives, peer pressure. And I think it really is the heart of the research that the people are driven by incentives, by their feelings, their behaviours, or their motivations are not as straightforward as one would expect. Their decision making is informed by many different factors. And we need to recognise at the coalface, these are people, people making decisions. And the more we understand how they make these decisions, the more we can create the right environments to ensure that they make the best decisions for the companies that they represent.

Juan de Castro: That is the analysis of some of the challenges, like what can organisations do about it? So you mentioned an example of algorithmic underwriting. Before you even get there, is it as simple as providing more visibility between what the technology or rating engine, whatever it is, is suggesting versus what's the ultimate decision done by the underwriter? Or is it not that simple? Help me here. How do you start addressing or at least providing visibility on those situations where it was obvious it was not going to be profitable, but you still wrote that business?

Tony Buckle: I think it's very important, Juan, to try and be objective and to the extent possible quantitative about behaviours rather than get into the world of psychology and very specific biases and that sort of thing. So sure, we've certainly found that people, for example, were much more influenced by peer pressure and the need to keep up with the competition than we had expected at the beginning. That was not something that we'd held out as being particularly relevant, but it was much more relevant. But there are ways, I think, of looking at behaviour that can be quantified, that may give very interesting insights into where we are, shall we say, in the market cycle. So at the moment, you can probably price most risks in a technical tool pretty quickly, right, without touching the sides too much. The market will bear the technical price the tool is going to require. So if you look at the speed at which you can fill out the pricing tool now versus perhaps the speed at which you could fill out the pricing tool before, how many discounts are being given at this particular moment versus how many discounts are given at another moment? These are very quantifiable factors and dimensions. They give a very interesting insight into what's actually going on in terms of the front line without needing to bring in the psychologists and go down into individual motivation. So I think we are looking at it, yes, behaviours are very, very important, but let's not stray too far away from things that management can look at from a quantifiable perspective and start asking some decent questions about that they can back up with data.

Juan de Castro: We've touched on a couple of factors, you mentioned peer pressure, for example. Also, in my experience in some of these organisations, there's also a logistical challenge, call it that way, which is in a soft cycle, you want to shrink your book. You also have an expense problem, right? I think the difference between theory where you would want to shrink and grow your book really fast, depending on market conditions and the reality of just having underwriting teams, is that it's not that easy to grow and shrink and you might not even want to do it. And I think this is one other area where you touch on your research, I think you refer to it as cost ratio driving appetite rather than loss ratio. Perhaps you want to touch on that one too.

John Carolin: We saw frequent and explicit evidence of top line targets being pushed to improve cost ratio. And I'm a financial analyst by background, I can see why the equity markets want to look at combined ratio and cost ratio as a key component of that when you're looking at aggregated data, external performance. But when you push that accountability down into underwriting for cost ratio, I think you need to be explicit or at least be clear that it needs to be a controllable cost. Every organisation should focus on cost. But if it's a myopic focus on cost ratio and a large part of that is amortisation or uncontrollable costs, you're really just formalising some cost bias. You're saying, let's make decisions today because of decisions that were made in the past that we can't change. You're telling people to do that. And it leads people to write. It's really throwing good money after bad. It leads people in a soft market to write business that is technically inadequate. And a few years down the road, that's going to be more claims coming through the loss ratio. So it's a very short-term fool's errand. And it's explicit that people behave this way. They write unprofitable top line business to move a ratio. And I think spitting that out and saying, these are controllable costs, this is the cost base that underwriting controls and it should have an eye on it, is at least one thing that people could do immediately. Not to say you shouldn't externally your cost ratios. As I said, I can see the value of that. And clearly, keeping costs under control is important. But how you do that, I think, can be improved.

Juan de Castro: And I guess there's also a tension between the short term and the midterm, right? Because I guess you are solving for growing the top line or at least not shrinking so that you can control a couple of percentage points of your expense ratio in this financial year, potentially at the cost of a long term, really having dug a deeper hole in the loss ratio, right? Is that how you look at it?

Tony Buckle: I think there's two points there, Juan. You have this sort of long term versus short term problem. And there is a bias within organisations to serve the short-term God at the expense of the long term God. So I'm going to hit my cost ratio target, and at the end of the day, the loss ratio can look after itself. It might even be somebody else's problem by then. So it's unfortunate, but with the long tail lines, that is a factor. And obviously, we've now got in the industry and widely broadcast the problems of the liability portfolios that were written in the midteens. So they're coming home with a vengeance now. I think it also requires us to think now, about how we prepare for the next soft market. And I think one of the things that we need to be thinking about now is how to scale up now and take advantage of the market now without creating my cost problem for the next soft market, right? So thinking about how can I scale my business now and not maybe hire a ton of additional underwriters, which we know in some markets are effectively semi-fixed costs, right, because of the social programs that exist around them. So can I somehow increase the productivity of my front end to enable that, which might be the sort of solutions that Cytora indeed has, you know, better triage, better augmentation of risk so I can make quicker decisions, making sure that I'm focusing and prioritising my stack in the most effective way possible, looking at the rejected risks and maybe farming that in a very productive way as well. There are lots of different ways that we can angle this. But at the same time, I think the key thing is, can I do that without adding a whole load more underwriters? Because unfortunately, once you've got more mouths to feed, you've basically created another problem for you next time around. And that problem is very real. And in a people business, where the majority of costs are the people's costs, that's something you really have to keep an eye on.

Juan de Castro: What you're saying is almost, before the soft cycle hits, you need to create infrastructure operating model, technology, everything that allows you to flex without significantly impacting your cost per piece, which is something that, especially large commercial is something that often carriers prioritise less so, I would say. I think what you're saying is that it is a huge enabler for then when the soft market hits, being able to be more disciplined in what type of risks you write, allowing you to eventually even shrink if needed without having to impact your cost base. And I think those concepts of driving productivity are obviously fundamental in that regard.

Tony Buckle: That's right. And I think it even goes as far as establishing now, what are your walkaway points going to be, objectively, when the market softens? And if you try to define them after you've written your plan for the year, you have a contradiction in the organisation. So trying to define those things in the fog of war, as John would put it, is very, very difficult. You need to define them now and get your stakeholders aligned to them. It's not just a question of, is management aligned to this? Are my underwriters aligned to this? It's often, are my investors aligned to this? This is what we are going to do. And make sure everyone understands that is the strategy, and that will enable you to navigate the soft cycle better than you've done it before. I think those sorts of discussions are very, very value-adding now, well ahead of the, shall I say, when reality bites.

Juan de Castro: One other area that those of us who are closer to carriers in the value chain are, I think everybody loves blaming the broker. I am sure that the brokers love blaming the carriers and the carriers, the reinsurers. This is like a funny game always. But one of the points you make also in the research is something along the lines of don't blame the brokers, which I think it was also a fascinating insight. Can you expand on that area too, please?

Tony Buckle: As you say, there is this sort of strange tension in the market and a desire to externalise fault. So it becomes easy to sort of blame each other. I think the finding we have is that the brokers were under pressure during the soft market. They're under competitive pressure. They need to try and bring the clients with them. They have a duty to get the best possible terms. However, the client defines that, by the way. They have the duty to get the best possible terms they can for their clients. And if the market is plummeting in terms of rates, broadening in terms of conditions and so on and so forth, it's very difficult for them to guess ahead of time where the market will be when the risk sets. So that is a very difficult situation for them to navigate to. It is not up to the broker to look after the carriers. The carriers have a duty to look after themselves and their capital. It is the broker's duty to get the best possible terms and conditions they can for their clients. And I think sometimes that understanding is lost and we end up sort of having rather strange conversations around who is responsible for things. Well, looking after your own area, I would suggest, is a better way to do it. And take accountability. Your own performance is a better place to start and a better principle to work from.

Juan de Castro: And it's very clear in the value chain who is responsible for what, right? So I think that's what you're highlighting.

John Carolin: Yeah, maybe to add to that, as Tony mentioned, sort of overall sentiment was brokers work for the client. They were doing the best job that they could for the client. Cordial and professional was the feedback. But there was this observation made a few times that during the soft market, yes, the brokers have a lot of leverage. But really, with all the pressures that underwriters are focusing on or feeling, they can expect that they will feel pressure from the broker. That's normal. But if you're a very hungry underwriter in a soft market, if you're chasing premiums, the brokers will find you. As someone referred to it, they will find the well-oiled hinged door and they will be going through it. They will find the people that want the risks and place them. So I think looking at all of those other pressures, one needs to be aware of if you've got a young, ambitious, or hungry underwriter, they potentially do get taken advantage of. And I think organisations need to be aware of that, particularly as the market becomes softer.

Juan de Castro: I think broadly, as you said, it's just an acknowledgement of carriers, brokers, reinsurers, each playing a very needed role in the value chain. And in different cycles, the power is on a different side of the value chain. So in a hard market, it's typically the insurers who have more power versus the brokers, in a soft market the other way around. And that is just dynamics in the market. But I definitely think blaming each other is not helpful. It doesn't really drive anywhere.

Tony Buckle: It went as far as getting feedback from some of the interviewees, which again was just really insightful for us, where people had been in trepidation about having to go to their clients and say, we cannot continue as we've been going on. We just can't do this deal at these terms and conditions. We're going to have to harden them. And the client’s response being, well, we've been waiting for this for several years. We know it's not sustainable on your side. Don't blame us for taking the terms you've offered us, but we've been waiting for you to tell us that. So I think there is an awareness amongst the broken client community of the need for the partnership to work for all parties. But the feeding frenzy that happens sometimes just grabs hold of the market in such a way the real interests of the clients and the brokers are, to some extent, just ignored. People are actually trying to second guess what the competition is going to do, not what the client needs.

Juan de Castro: Yeah, we should never forget at the end, all these are services we provide to the client. So it should be in the client's best interest always. So across many of the topics we've touched on so far, we've always had this transversal angle of culture, management communication, setting the right expectations, setting the right strategy. Let's focus specifically on that area for a minute. And what are your findings and views on that?

John Carolin: Management communications was a very interesting one because we saw divergent practice around how the messages from management and the implications for those were very interesting. So when I say divergent practice, there were underwriters we spoke to where there was a clear and consistent message in one of two directions. One was you need to write profitable business. Discipline is important. Don't chase the premiums. And then we even had cases where there was a potential trade sale taking place and communication from management was that we needed to go after the premium. But the message was clear. By and large, what we had was a lot of unclear messages. So what we'll call ambiguity, you need to hit your premium targets and it needs to be profitable. And when pushed for which one of those two it is, it was to go and find the good deals. And almost by its nature in a soft market, the only way to grow premiums is to take on more risk at poorer terms. So a lot of underwriters are really struggling with this ambiguity and how to act. And even instances of discord between what was being said and how people were acting. So we heard feedback from underwriters saying, yes, management preached a message of discipline, but every week we would have production meetings where we would celebrate new deals. And after many weeks of not being celebrated, the message says one thing, but the way people are acting is very difficult. You don't want to be that person who never gets celebrated. So I think the last point I'll make on this was the survey work that we did at the beginning. There was a lot of thought that went into the questions we asked because we wanted to paint a very broad picture. But at the end of it, we did a sweep across the data or correlations between questions that we hadn't expected. One of the ones that jumped out at us was this correlation between management communications focused on disciplined underwriting and trust in management. When underwriters saw management talking about discipline, they had a high trust in management. Those two were very highly correlated, and those weren't a paired set of questions that we'd specifically put in there. And so I do think it's one of the more interesting findings that jumps out.

Juan de Castro: I find this type of finding really fascinating because it's often perceived internally in insurance companies that when a book is loss making in a soft cycle, the underwriters didn't pick the right term. So the underwriters didn't choose the right risk. And I think what you're highlighting here is that the senior execs on the board should not complain because it almost starts with that clarity of strategy, messaging, consistency of what you celebrate, to your point about celebrating new risk. Do you celebrate binding new large risks or do you celebrate the discipline in picking the right terms or the right risk? I think this is something really to reflect on. What's your experience, Tony, in these regards?

Tony Buckle: I think I'll add another couple of things. I mean, one of the things that again jumped off the page for us was engagement scores. So a lot of companies have these employee engagement surveys now. And we asked people about how was your engagement at the time, thinking we were going to get, frankly, diabolical scores on the grounds that they must have been really frustrated, they were under all this pressure. And they actually said, actually, we were engaged at the time. We were responding that we were engaged at the time. So you had this sort of contradiction there at the heart, which was, if you were looking at the engagement survey to give you insight into underwriter mood, that wasn't giving you really that strong an indicator at all. In fact, it was almost giving you a contradiction, because people were saying they were very engaged. That itself is a problem. If people are saying, well, let's look at my engagement scores, and if they're very low, that would indicate that it's a tough trading environment out there, and people are very frustrated. If they're very high, oh, they're finding opportunities to do deals. They are motivated. It's important that we get a clear line of sight. And I think underwriters are willing to provide insights into how they really feel. We've had a good amount of feedback or statistically strong feedback that they did inform where they were. They did say, if we grow now, we're going to lose money, even down to deal specifics. If we write this deal, it will lose money, and being basically instructed that I've got the message, we're doing that. And I think that's very interesting. So you have these sort of contradictory themes in our view that what was going on in terms of practice was incredibly demotivating, but the scores that they were providing for the employee engagement surveys were very high.

Juan de Castro: It's really fascinating, all these insights from this research. I'm almost thinking we should title these episodes like the soft side of the soft cycle. I think what you are bringing to life is something that the industry really thought for many years is about just underwriters, as I said earlier, picking the right risk, but it's so much more nuanced. It's so much more related to the culture, to the message, to the way an organisation is managed. In the last few minutes, I would really love to just almost zoom out again and get any final takeaways, anything else we have not touched on.

Tony Buckle: So to be honest, Juan, I mean, there's a lot of detail. We've gone into everything from accountability through to operating models, the specific metrics, how do you balance internal metrics versus external metrics. So there's a lot more detail we could go into. I think that the key thing for me, and I recognise I'm repeating something I said earlier in the call, is I think this idea that soft is somehow not quantifiable, I would challenge that. Did underwriters know that the market was soft? The classic judgement request. We've been able to statistically prove that the market knew that it was soft. The underwriters knew the market was soft. They knew the terms and conditions were deteriorating. And, you know, once you get into sort of quantitative stuff, you're getting into what gets measured gets done territory. So you can actually measure it and you can actually make better metrics and you can target better things and better outcomes. So I don't want people coming away from this thinking this is purely psychology and behavioural analysis. It is very quantifiable what we can do, not only in terms of demonstrating where we are at a particular moment in time, but also in terms of the improvements that can be made.

John Carolin: We have looked at lessons learned. We've taken suggestions from people. We've looked at varied practices. But I think when you really zoom out, if you think underwriters are making their best loss picks because they have valid, accurate, and complete pictures of the risk, then unfortunately, we've shown that you're wrong. And I think we knew that. That's why you do portfolio analysis. The sum of a portfolio is more than the sum of its parts. You know that you can squeeze additional performance out of a portfolio when you look at it from a portfolio perspective. I think with the lens that we've been looking at this, you can start to measure things that do matter and drive decision-making at the coalface and can shape the environment to improve that and ultimately manage the cycle more effectively. Because as Tony said, I have to use the word now, in the fog of war, you can't make the best decisions. If you can't take this knowledge and do something about it, we'll just find ourselves back in the same place. I think history shows us that markets swing from boom to bust, and that doesn't build sustainable markets for any of the participants. So I think there's a wonderful opportunity to look at this information and reshape the environment in which underwriting decisions are taken.

Juan de Castro: Well, Tony, John, it's been a fascinating chat. I've really enjoyed the angle that you've brought from this piece of research. To me, the key takeaway is successfully navigating a soft market is both about what you write, as well as how you manage the organisation, we talked about culture management, and the fact that you can measure both. I think it's a very positive message. A soft market is not something that just happens and you're a passive actor. You can actively manage it and be successful. I think that is what you're doing in the industry and helping carriers globally in this transition. Thank you so much for joining me today. It's been fascinating.

Tony Buckle: Thank you very much, Juan. Always a pleasure.

John Carolin: Our pleasure, Juan. Our pleasure.

Juan de Castro: Making Risk Flow is brought to you by Cytora. If you enjoy this podcast, consider subscribing to Making Risk Flow in Apple Podcasts, Spotify, or wherever you get your podcasts so you never miss an episode. To find out more about Cytora, visit cytora.com. Thanks for joining me. See you next time.