Beazley plc (BEZ.L) Earnings Call Transcript & Summary

Hello, everyone, and thank you for your patience. Welcome to the Beazley's Year-End 2021 Results Call. My name is Daisy, and I'll be coordinating this call. [Operator Instructions] I'll now hand over to your host, Adrian Cox, the CEO, to begin. So Adrian, please go ahead.

Thank you, Daisy, and good morning, everyone. Welcome to the presentation of Beazley's 2021 results. I'm here with Sally Lake, our CFO; Bob Quane, our CUO, having successfully made it across the Atlantic this last weekend. I'll take you through the overview and dividend policy. Sally will then discuss the financials and share the usual metrics. And following that, Bob will delve into a little more detail on the underwriting, including drivers of growth and thoughts on climate change and he may even mention cyber. After that, I'll present our outlook for the coming year and we'll go on to Q&A. On to the overview then. We're pleased with these results. Whilst we're not yet back to the heady heights of 2012 to '16, we have generated the largest profits in our history. And it's a combination of underwriting margin that's returning back to the sort of levels that we target and a reasonable investment income in what was quite a difficult year, particularly the last quarter. Our combined ratio was 93% and for us that's the metric that really matters. And we are working very hard to try to continue to reduce that further this year. Our combined ratio includes all costs, though it's a very transparent measure of the real margin in our business. It's good to deliver on our promises and this is slightly better than the mid-90%s to which we guided in the half year and the third quarter IMS, and it's a reflection of lower than budget for cat activity in the fourth quarter. Again, we expect this number to improve given both what we've done over the last 4 years on the underwriting side and our expectations for market conditions looking forward. So profits then of just under $370 million and a return on equity of 16%, which is respectable and given our expectations for the combined ratio should also continue to improve. Of course, the reason why our ROE isn't even higher is given those record profits is that we're somewhat larger now. As a comparison, our gross written premium in 2016 was less than half of what it was in 2021. And there are 3 thoughts, I think, that spring from that. Firstly, the combination of the specialty products that we have and the domestic and wholesale platforms that we have to sell them on really does enable strong compound year-on-year growth, even allowing for the prudent cycle management that we pride ourselves on. And Bob will discuss these drivers a little more. Secondly, the bigger asset base that this generates should yield greater investment income going forward. And lastly, after these last few years, we're now a Tier 1 market for our brokers across our wider proportion of our products, which is absolutely in line with our ambition and makes us a stronger business. We don't pursue growth for its own sake. We've always obsessed about underwriting profit, but it is exciting, I think, that we're in a position with multiple opportunities for long-term profitable growth. Talking of which, we grew 30% last year, 24% of that was driven by rate. But I'm satisfied we did manage to put some exposure on the books as we've been talking about doing that for a little while now. It's the right time to do that and we'll continue to do so this year. Good reserve releases across the board of just over $200 million. Again, with a bigger engine now, these numbers should continue to grow, and Sally will go into that in a little more detail in a few moments. Our surplus capital is slightly in excess of our target range, third range of 15% to 25%, although that will go back down to 22% post-dividend. And I would note that the 27% does contemplate the funding of this year's business plan. And lastly, as we were indicating last year, we will be paying a dividend of 12.9p, which is up 5% from the 12.3p we last paid in 2019. And I have a few more thoughts on that in a second. Just before I move on to the next slide, I'd like to announce that we will be holding a Capital Markets Day in May this year. There's a lot of stuff to talk about, a lot of exciting things going on here, too much really to share in a year-end presentation like this. We thought we'd like the opportunity to spend some more time with you, showing a bit more detail on topics like our cyber infrastructure, IFRS 17 and our new digital business. So if we look back at the 5-year trends, it is evident I think that this performance is a better one than we've produced in a little while. And what's pleasing about it is that we've done so in a period of heightened risk and claims activity. So it's not just a benign claims environment that's driven this result. And we've been working very hard on that.

It appears we have lost the audio from the speaker. Please can you check you aren't muted?

No, I'm not muted.

Apologies, ladies and gentlemen. Please bear with us as we reconnect the speakers.

Okay. Good morning again, everyone. I do apologize for the technical difficulties we appear to be having. I think we lost the audio for a while. We're getting some techs in saying that the slides are jumping around. So I do apologize if you're having some difficulties. I'm going to start again in Slide 6, which shows strong premium growth and a return to profitability. And I was just saying, I think before we lost the audio, that you can see from the slides that last year's performance is a better one than we produced in a little while, and we're pleased we've done it in a period of heightened claims activity. It's not just a benign claims environment that drove the combined ratio because we've been working quite hard on this for a few years, and I'm pleased to see those efforts start to bear fruit. I'm going to move on to our dividend policy. I discussed some of the principles around this at the half year, but I thought it would be worthwhile going through them in a little more detail now they've been finished and signed off. We have retained a progressive dividend and it remains a strategy that will return excess capital to shareholders. So if we can't use the capital for profitable growth, we will return it to shareholders either through the progressive dividend or through a special and we are going to move from a twice-yearly payment to an annual one. And so what is our thinking behind that? Well, our primary aim is to invest in future profitable growth for the business. And the primary use of retained earnings is to do just that. Our planning cycle is an annual one and ends in December, and therefore in December, we'll have the optimal view of what the business needs are, what the investment opportunities are and what our capital availability is, which means we'll be able to make the most informed decision about dividends in January. So for us, a single annual payment seems to make great sense. We chose to retain a policy that is progressive because we believe in the financial discipline that that drives. And as I said, we continue to believe that if there isn't a good opportunity to invest our retained earnings in the business, we will return it to the shareholders either via the progressive dividend or through a special. However, fortunately, as I think I've been communicating, there is lots of opportunity for us to invest in the business at the moment, and we hope and believe that that opportunity will persist. So with that, I'll hand over to Sally to present the financials.

Thank you, Adrian, and good morning, everyone. My name is Sally Lake, and I'm the FD here at Beazley. I am hopeful that you can see the slides. There is some confusion that they may be jumping around. So just to let you know if that is happening, we will be putting them on the website, but please do bear with us if that is happening. Okay? So I'm going to take you through the numbers in a little bit more detail, and then I'll do a quick focus on investments, followed by reserving with ending on capital before handing over to introduce Bob. So Adrian gave a lot of detail already, including the impressive growth that we've achieved over the past 12 months, slightly less when we look after reinsurance as we continue to purchase some additional reinsurance in the areas where we are seeing growth significantly. It is pleasing to see our claims ratio also reduce significantly from its high of 2020, which was caused by our COVID-19 exposure. And as we continue to grow, we are also seeing a steady reduction in our expense ratio where we are able to take advantage of some operational leverage. At its height at a very different point in the market, this ratio was 41%. Whilst it's pleasing to see this reduction, we are also continuing to invest to ensure that we have the right systems, processes and technology in places so we are well-placed to maintain our expense ratio at lower levels throughout the cycle. So on to investments, and the first thing I want to note is the overall investment balance has grown by over $1 billion in the past year, which is just another reflection of the overall level of growth that we are currently seeing. The investment balance at the end of 2021 was just shy of $8 billion. As ever, our portfolio makeup remains broadly consistent with the majority of our assets being sovereign and investment-grade debt and the remainder being held in capital growth assets. Similar to last year, we have continued to move from investment-grade sovereign debt. This is a deliberate thing that we are doing as with falling credit spreads, the attractiveness of the investment-grade bonds has reduced. So let's have a look at the return that we saw. And I think given the backdrop of the investment market, particularly in the second half of 2021, the team have done a terrific job with a return of 1.6% in 2021. This was achieved through a number of management actions during the year. And whilst our equity exposures ended where they begun at around 3% of our overall investments, it's interesting to note that at points during the year, there was high 6% as we took advantage of the strong rallies that we saw. We've also added some exposure to inflation-linked bonds within our sovereign debt portfolio, and this helped returns as inflation concerns have been growing. We also chose to reduce the duration of our fixed-income exposures as yields rose later in the year, which helped reduce any capital losses. First weeks of 2022 have brought headwinds to our investment returns with fixed income yields rising quickly and equity markets reversing some of their recent gains. We have kept our duration below neutral and reduced our equity exposure, which has helped to reduce the capital losses. We continue to hold a diverse makeup of capital growth assets. Also whilst the increase in yield just give a short negative impact on a mark-to-market basis, it does improve outlook for future returns. As ever, we continue to remain within our risk appetite and use our management actions where appropriate to both protect the portfolio and search for ways to add return. Now on to reserves. The reserve releases are usually a large contributor to the overall underwriting profit here at Beazley. And when I took over as FD in 2019, I was not happy with the reserve release position. And we've been very, very focused on improving this picture for a lot of hard work around our underwriting actions. And I'm extremely happy to see the outcome continuing to improve this year and all the hard work beginning to pay off. Notably, for the first time since 2017, all divisions have seen redundancy in their reserves, and they've been able to contribute to a reserve release of over $200 million. It's also pleasing to see the line which shows the overall reserve release as a percentage of net earned premium on a nice upward trajectory. This year's result is around 7%, but this is still below previous good years, which yielded releases in excess of 10%. It's important to note here that particularly in times of significant growth like the one we're seeing at the moment, there will always be a lag between net growth and the release of reserves within this business. This is because we wait until we have a good degree of certainty around the claims we are reserving against before we start to release the margin that is emerging. And this is particularly true of our liability business within specialty lines and [ SCIEX ]. There are no -- let's look at the level of reserves that are on our balance sheet. And this is a graph that we've been showing for many years now. It compares the level of reserves on our balance sheet to a consistent actuarial measure, which in itself has a level of prudence within it. We have a preferred range of above this measure of 5% to 10%. And showing this graph aims to demonstrate that we are approaching our reserving consistently and prudently over time. Now as I mentioned on the previous slide, the amount of reserves we have released has increased this year, but you can see that this wasn't achieved by reducing any strength within our reserving level on our balance sheet. This remains very consistent with the previous few years. It's worth mentioning at this point that we have IFRS 17 coming into effect in 2023, and we will be looking at how this affects our disclosures, including those in our reserving levels, and there will be more to come on this in future updates. Finally, on to capital. So nothing new to report on resources with a mix of Tier 2 debt and the banking facility along with our equity making of the funding. As ever, we measure our capital requirements in a number of ways, but we continue to choose to speak around the lowest ECR, which is the most onerous capital requirement and much more prudent than standard Solvency II measures. It considers the business to ultimate rather than over a 1-year time horizon, and it also has a further 35% uplift. As Adrian already mentioned, we've already modeled the business that we are planning to write to the end of 2022 within these calculations above. In addition to the Lloyd's measure, we also show our U.S. insurance company, which is subject to a risk-based capital regime. Now you can see that the requirements have increased over the year, but they have not grown in line with our premium growth, and they are lower. And this is because we are starting to see positive impact of the rating environment beginning to benefit these requirements. We're also continuing to focus on writing business on the most capital-efficient platforms wherever possible. Now where does this leave us? So we end the year at 27%, which is just above our preferred range of 15% to 25%. And this means 2 things. Firstly, it leaves us well-placed to reinstate the dividend that Adrian has already mentioned. And after paying this, the surplus will move to 22%. Secondly, it means that if there are indeed further opportunities over and above those that we've already planned for in 2022, we have the capital available to take hold of those opportunities. And now over to Bob.

Thanks, Sally. Hello, everyone. My name is Bob Quane and I'm the CUO of Beazley. As I'm new to Beazley, I thought I would start by introducing myself. I've been in the insurance industry for more than 30 years. In addition to the United States, I worked in Europe for 13 years. And my last 12 years in New York, I've been in global roles. I've led personal lines and commercial line products and consider myself a deep generalist with strong understanding of individual products. I am a problem solver, and I believe I will bring a fresh perspective to Beazley. Today, I'll talk about the growth across all of our products, give an update on natural catastrophe, discuss our approach to climate change and give an update on cyber. We often speak about cyber, but I think it's important to let you know about our exciting growth across our other divisions. We can grow across a broad range of products, which gives us long-term opportunity for growth. We actively cycle manage and grow in the hard markets and show restraint in the soft markets. In the next 2 slides, I will compare gross premium growth against gross premium growth adjusted for rates, which is a proxy for exposure. Looking first at cyber and executive risk, growth is driven by cyber until 2019. And the D&O market heightened -- hardened, and we shifted the growth to D&O. And then in 2021, while we had dramatic premium growth in cyber, we reduced exposure at that time. Turning to specialty lines. We've experienced a broad market positive rate change from 2019, and we have demonstrated a consistent exposure growth across those products. Moving to marine impact. In 2020, we had an exposure reduced due to the collapse in demand for event cancellation insurance. Other than that, we've had consistent exposure growth. Finally, we have property and reinsurance. We've had exposure reduction since 2017, reflecting a poor risk-reward environment as losses have increased due to the impact of climate change. As you can see from these charts, we have grown across almost all of our products at Beazley, except for property. The story is much bigger than just cyber and it illustrates that we are well-positioned for balanced growth. We will continue to move cautiously with property until we're confident that we're getting the right return for the risk. This slide shows how we managed our risk appetite for natural catastrophe. As a percentage of net earned premium, the mild loss at the 1 in 250 has been decreasing. You would see a similar trend on a gross basis as we have not made any material modifications to our reinsurance structure. In absolute terms, the 1 in 250 has been growing modestly, but the 1 in 10 has been shrinking since 2019. This will assist us in managing volatility of our earnings. This approach reflects relative deterioration in risk/reward and active cycle management. We will continue to move with caution and manage risk appetite carefully as we navigate through the challenge of underwriting for climate change. As we work to get our arms around climate change, we have reduced exposure in property. I've already covered this in previous slides, and we will continue to move with caution until we are comfortable. We are building capabilities to get our arms around the challenge. In terms of people and expertise, we are strengthening our pricing and cap byline teams, and we're hiring a financial climate risk expert. For predictive tools, we are improving our raters, our underwriting tools and cat models. We are also exploring an external tool that will allow us to understand how climate change impacts the physical assets of our clients out to 2050. To account for climate change, we will add a forward-looking view of risk to our predictive tools. We're making risk disaster scenario more robust for climate change, such as creating a scenario for green washing for liability. And we will strengthen our client engagement to validate their commitment to ESG and understand their path to transition. Climate change complicates underwriting and pricing, which will make products like property less commoditized. We believe this plays to our strength. After we get our arms around it, we will look to grow property. This is an update to the graph and numbers we shared at the interim results in July '21. It shows that the frequency trend continued in the second half of '21. We implemented stronger underwriting standards in October '20 and you can see the powerful impact it has had in reducing frequency on both a policy count and a premium basis. Comparing '21 to 2019, our exposure has reduced. In fact, '21 is on the same standard level as 2018. We believe we have better risk as evidenced by the materially lower frequency, and we have significantly more premium. Before I hand it back to Adrian, I'd like to say more about our approach to cyber. Firstly, the downward trend in claims is driven by underwriting actions taken in October of 2020. This includes a range of tools, and we continue to evolve and refine them as we move forward. Secondly, climate conditions remain favorable. Demand for cyber products is increasing. Insurers are strengthening their underwriting and reducing their exposure and reinsurers are also tightening their position. In the second half of last year, we saw rates more than double, and this has continued in early 2022. Thirdly, the investment in our cyber ecosystem are on track and continue to evolve. Some highlights include Beazley cyber council, panel of global experts that advises on cyber threats. Warnings of ransomware attacks are delivered from threat intelligence and data partnerships to give our clients early notice of the risk. Client service portals. These helped clients manage the impact they suffer -- if they suffer a cyber attack. The success of our cyber ecosystem responds directly to the final and perhaps the most important point. We are seeing strong and rising demands from our clients for more risk management support to improve their overall cyber resilience to help them avoid an attack in the first place or to mitigate and manage the problem should the worst happen. In summary, we believe we are well-positioned to remain a leader in cyber as we move forward. With that, I will turn it back to Adrian, who will talk about Beazley's outlook.

Thanks, Bob. Well, I think we've discovered the technical problem. We don't have enough screens in front of us. I think we're controlling them well enough. So I'll switch this off and Sally will control the slides. So to conclude then, we are feeling quite positive about the upcoming year and beyond. Yes, we remain in a heightened risk environment, but the market is allowing us to do our underwriting due diligence and to price for us, obviously. And we can live with that. So we're planning for continued growth well in excess of our long-term 5% to 10% average boosted by some positive rate change. I'm not sure we'll reach the quite heroic 30% that we achieved last year, but it should be comfortably in double digits. And as I mentioned earlier, we will be looking to add additional exposure to the books this year. We have the capabilities to capture the opportunity that this market presents and I expect growth and opportunity across our wholesale platforms in London, Singapore and Miami and our domestic platforms in North America and Europe. We will be growing our cyber business and in premium terms, it will represent more than 15% of our business this year. However, when we rebase those premiums back to 2020 terms by taking out the rate change, it is in line, it will remain in line with the diversification principles that we believe in. We do remain firm adherence to the benefits of a well-diversified business, and then exposure terms of business will remain that way this year. Given all the above then, we're expecting a combined ratio of around 90%, given average claims, which is more in line with our long-term experience. The investment yield that we show at the end of 2021 has risen a little, as Sally has said. And lastly, I'd like to reiterate the invitation for our Capital Markets Day on the 18 of May. It will be the hottest ticket in town. So do book early to avoid disappointment. And with that, I'll hand over to Q&A.

[Operator Instructions] Our first question is from Freya Kong from Bank of America.

Two questions if I can, please. Firstly, if I normalize for above-average cat losses this year, I get to around a core of around 90%. How does that square with your effectively flat guidance for 2022? Could you walk us through the moving parts or assumptions within this? And secondly, just on higher inflation, where are you closely monitoring inflation and which parts of your book are most vulnerable to potential wage inflation?

Okay. I'll take the second question first, if I may. So we moved our loadings this year from fears of COVID claims and recession to inflation. So we have loaded our loss picks for that. Slightly different parts of the book are infected by inflation and were recession and COVID. And really, it falls into 2 broad buckets: one, social inflation that we've been talking about for a while, which we continue to load for; and then more retail price inflation, which affects more of the asset classes. So we have adjusted our modeling and our pricing and our loss picks accordingly. Inflation in and of itself doesn't create a problem for us as long as we -- it is unexpected changes to inflation causes an issue. Once we've factored it into our loss picks and our pricing, so on and so forth, we're generally able to cope with it fairly well. As far as wage inflation is concerned, we did see a slight rise in attrition levels last year, mainly because we think they've been so low previously. During the first few lockdowns, very few people left because people won't do much at all. We're still very comfortable with our attrition rates and with our -- the culture and the feeling of the organization. And whilst we are seeing wage inflation a little, it still remains very much under control, I think. Sally, do you want to talk us through the combined ratio?

Okay. I'll do the next one. Can -- well, can you maybe shut your iPad because apparently, we keep going back to investments, which I don't -- we don't know what's happened to the slides. It's all very exciting, but focus on our words, not our slides at the moment, guys. So no, it's a good point around an average cat. So if I think about what happened during the year, we started 2021 saying that we were expecting the low 90%s. We ended at 93%. We can talk for hours about whether 93% is a low or a mid, depending on your preferences. I think that in an average cat this -- the past year 2021 would have been probably quite a good low 90%s, but I definitely see a delta between an average cut in 2021, our expectation in 2020. And so I do see some improvement. But you're right, Freya, that with an average caps, and I think the way we guided during the year was that we've seen claims as expected at the low 90%s, slightly higher than average cuts led us to 93%. I'm not going to comment on what that is. And then next year, with an average cut we're expecting over 90%. So there is not a huge difference, but a definite improvement between how we entered 2021 and how we're entering 2022.

And that was driven by the fact that it takes us a while to start to release our long-term reserves. And that's why the delta is perhaps slightly less than you might have thought.

Okay. Great. So I guess, yes, just thinking about the rate improvements that you've got 24% in 2021, and it should be quite strong in 2022. How should we think about that earnings through into the combined ratio net of claims inflation, any other assumption changes?

I think it feeds into the combined ratio of around 90% that we are -- that we've guided to, Freya. The result this year is a combination of what we do this year and what happens to previous years, and it's the latter that has the bigger influence on our combined ratio, and that's what we are hoping to guide towards. That makes sense?

So I guess going forward, Freya, if you're thinking what should you be expecting beyond 2022, you're right that if we're writing business at a higher rate in 2022 than 2021, you would see all else being equal improvement on our short-tail book, of which I include cyber, they normally take 2 years to come through and on a longer tail book, which is generally most of SL and the X of SCIEX, I would suggest that would start taking positive movements after 3 years. But obviously, we've been seeing those improvements for a number of years now. But given that we are continuing to see positive rate improvements, and you should see that incrementally coming through beyond the end of 2022.

Our next question comes from Kamran Hossain from JPMorgan.

So a few questions. I guess just following up on Freya's point. I guess thinking about reserve releases and the combined ratio going forward, should we think about reserve releases both growing in an absolute amount, but also kind of a greater amount relative to NEP going forward given that there is this kind of unwind of prudence over time? The second question is just on capital position and the use of reinsurance. Your capital position came in above the top of your range. I guess after dividend, it's now within the range. But given that strong position, do you intend to keep a little bit more of the business this year or what are the plans going forward? And the third question, still on capital. How should -- I mean, you've got this new dividend policy today. How should we think about excess capital? Is it anything over 25% or is it going to be far more baked than that?

Okay. Let's go through those in turn. First question, the answer is yes, Kamran. So as Sally pointed out in a chart that we may be able to show you.

It's the right computer.

The computer. That one. Our reserve releases as a percentage of NEP are up, but they're still lower than they were when we were throwing out the 90% combined ratio more regularly. And if we do perform as we expected, those are the trends that we should see again. And as I mentioned, we're a much bigger business now than we were. And so the dollars will increase also if that is correct. So the answer to your first question is yes.

Yes. I think just to follow up on that. I think the 3 things that are going to drive the underwriting profit, exactly what Adrian said around reserve releases. What happens in terms of net cats versus the average cat margin that we're holding and how we open our reserves as well. And I note the third one is, generally, we do open consistently over time. However, given what's happened in cyber in the last couple of years, you will see that we've opened SCIEX slightly higher because of that. So that will also impact the underwriting profit. So there's -- it's not -- there's not a direct correlation between just the reserve release. It also is dependent on our opening position as well and clearly net cat as well.

No. Good. Second question around use of reinsurance, you're right, we have been using proportional reinsurance to support our growth areas where they are particularly high and capital-intensive. I think we will continue to use reinsurance. We are greedy, of course, to keep business on our own balance sheet where it makes sense. And part of that is growing our capital base. Part of that is optimizing the business mix, and we do both, and we want to make sure that net the business remains properly diversified and optimized. So all other things being equal, we will be buying slightly less reinsurance going forward, whilst we maintain that optimization as we grow retained earnings. But part of it also depends on what happens to demand, Kamran. And -- because if demand continues to increase as we have seen it, then the growth rates that we are able to achieve in those businesses will exceed the rate at which we're growing capital ourselves, and we'll continue to use the reinsurance to do so. So it's a relatively dynamic picture. But I think if what you're fundamentally asking is, as we grow bigger, do we want to keep more? The answer is yes, we do. And then lastly, your third question around excess capital. Yes, absolutely. If we have more capital than we can usefully invest in the business, we will continue to return that to shareholders as we have always done either through growing the progressive dividend or through a special dividend. We're very actively capital managing. We always have done and it's a good discipline have and one that we commit to, sort of as I was trying to intimate though in the presentation, there's lots of stuff going on at the moment, lots of opportunities. So we are hoping not to need to return capital to shareholders in the next couple of years at least because there's lots to invest in. But as and when that changes, and it's a cyclical business, then we absolutely will do.

And I think in the past, when we've returned, it's not necessarily been above a certain number necessarily because it tended to be how we feel about our prospects at the time as well and the other opportunities that are out there. So we've never necessarily always targeted a certain special [ db ] percentage after. It's a lot to do with how we're feeling about our prospects going forward. So unfortunately, we're not going to give you a number to model there because I really think it's dependent on the situation.

Our next question comes from Derald Goh from RBC.

I've got 3 questions, please. So the first one is just in terms of the growth levels. I mean I know you mentioned double digits, but can you be anything -- any more specific about that? I think you previously alluded to a mid-teens net growth for 2022. How has that changed? And also within that, what level of rates have you kind of assumed? And second question is on cyber. So I mean, on one hand, I hear that you're quite constructive around the market conditions. But then it sounds like you're going to keep your exposure flat. So maybe how do I square the 2 elements there? And maybe also explaining a bit more around the capital reallocation, what have you done differently from before? And if the reduction in cat risk frees up any capital at all? Thirdly, it's just around reinsurance purchases. The contingent quota share that you got last year, has that been fully utilized yet? And if I heard you correctly in your introductory remarks, you said that you bought additional reinsurance this year. So maybe you could elaborate on what those new reinsurance programs are, please?

So growth levels, yes, I mean net growth levels of around the same that we achieved this year roughly and sort of the same mixture of rate change in exposure, I think, roughly. It's more of a mixed environment in terms of rate change than perhaps it was a year ago and it's quite difficult to predict where some rates will go across various bits of the business. But overall, we're thinking it will produce same sort of result that it did this year. Cyber, we're expecting rate increases to continue there. So certainly, a chunk of the premium growth this year will be rate. We are hoping to put a bit of exposure back on the books this year and start to grow our policy count again because you can see it's back down to where it was a 1/3 of the way through 2018 currently and the ambitions we have are to get that client base back again once they've invested appropriately in network security. But we will do that according to what's happening in the marketplace and the risk/reward we can get and the continued growing confidence we have that the re-underwriting and the support that the cyber ecosystem has provided us with the necessary confidence to put exposure back on the book. So as we learn more, we will continue to act. And like much of the plan actually this year, it's going to be quite dynamic and dependent upon what's happening to the opportunity and how confident we are with the results we have are things that we can continue. On capital reallocation, this is something that we do actively each year, make sure that we optimize the allocation of capital net, we get the best mixture of volatility, contained volatility and return on equity. And we have been moving capital around. I think what we were trying to show with the slide showing our 1 in 250 model net cat appetite against net written premium is that we've been gradually allocating less capital proportionately towards that as the risk/reward has been deteriorating and their increasing payrolls that we're not getting paid for. If we can start getting paid for taking those risks again, we'll absolutely move capital back the other way. And that's our aim. And I think one of the things that Bob was saying was climate change does present opportunity I think for insurance companies that can get their hands around it well and provide the right solutions and hopefully, there'll be opportunity to help our clients and grow our property book again because it's a smaller part of our business than it was 10 years ago. Reinsurance -- the reinsurance question, have we used the contingent reinsurance that we talked about last year? Yes, a bit of it, not all of it. There's some to go. There is plenty of enthusiasm in the reinsurance market for the sort of partner reinsurance we buy, which we used to help us grow. And it's worked well for us, and I think it's working well for our reinsurance partners. Does that answer Derald?

Maybe just quick follow-ups. So firstly, were there any changes to your capital allocation towards systemic cyber? And secondly, on the reinsurance, were there any additional new programs that you added on this year?

Yes, we didn't report it, but it was reported in the press that we bought a quota share for our cyber book. We've always bought some proportional reinsurance on our cyber. We have bought more this year because our cyber book is growing. And I think that's the right thing for us to do. We always but -- as I say, we try to manage our net book so the business is optimized both in terms of ROE and volatility. And because cyber is growing that is what we've done. And are we allocating more towards systemic risk? Not particularly. We've always made sure that we spend a lot of time thinking about what the systemic risk is, what scenarios we are exposed to, what those mean to us and make sure that we manage our portfolio grows so that we can control them and that we manage our reinsurance hedges, whether the net outcome of them is acceptable to us, and we have continued to do that as the scenarios that we build change with the evolving threats and as the book changes. And one of the things that's actually happened this last 18 months or so is that as our policy count has reduced and our aggregate has reduced and some of the additional coverages that used to be bought when the prices were cheaper have gone away. Actually, in numerical terms, our systemic exposure has reduced although the threats continue to change and evolve.

Our next question comes from Faizan Lakhani from HSBC.

My first question is on the capital walk from half year to full year. If I'm correct understanding at the half way you had the 2022 growth in there. So the capital has moved 4 points. When I look at the capital requirements, they stayed broadly unchanged. And if we assume your combined ratio as guided from half year 95% to 93%, explains about 2 points of that. Can you explain where the remainder of the walk stems from? The second question is on reserves. The [ Cybex ] underwriting year '19/'20 continues to benefit from the aggregate excess loss program. How much protection is left on those 2 underwriting years? And I guess just as a sort of a second part of the second question, the gross loss ratio on the latest underwriting of Cybex has improved significantly since 2020. There still sits 4 points above sort of historical levels. From this, can we assume that more needs to be done to improve the Cybex performance? And my final question is on the absolute growth in your admin expense base. Obviously, we've got inflationary pressures. We've got possibly that you're reinvesting the business. How much growth can we expect in terms of absolute terms going forward?

So we've got 4 questions there. I can answer the first one really well, but probably not as well as you can. Why don't I pass over the capital bridge to you?

Why don't I do the first and the last and then you fill in the gap? Okay. So it's a really good question. I would add 2 things to the movement between half year and full year. Firstly, the capital surplus is done on a Solvency II basis rather than a GAAP basis, which is what our combined ratio is on. And what you see there is you see earlier recognition of the positive rate changes coming through than we would see on our GAAP as discussed earlier that when we get -- when we see positive pricing improvement -- I want to keep going. Let me just check. We've got rid of -- there're some background noise that we just ensure we get rid of before I start again. Okay. So I'll start again. The capital calculation because it's on a Solvency II basis, it will see the benefits of the positive rate change quicker than GAAP. So because it doesn't have a margin, as we showed in reserving as such. So actually, the positivity you see coming through on a GAAP isn't always -- the movements within the GAAP combined ratio aren't always exactly the same as the movements in the capital ratio because they're on a slightly different basis. And when you're seeing a positive rating environment Solvency II recognize that faster. That's one thing to note. The second thing to note that we have finalized between half year and full year is that we set up a captive in the U.S. to do a loss portfolio transfer of our admitted business last year, we transferred our '18 and prior accident years from the end of 2020. We transferred the '19 accident year in the end of 2021, and that will also have a positive benefit on the U.S. RBC calculation as well. So there are the other things to add to the mass, which I don't disagree with what -- how you went about the mass, but there are the other things to think about. That was on the capital. In terms of...

I just want to follow up on that quickly. So first, in terms of the captive change, yes, captive number hasn't changed all that much between half and full year. And then in terms of recognition rates, has your view on rates changed since half year? Or is this the sort of future premium that's coming through in the calculation that has better rate?

So in terms of the captive, what will have happened is that any -- is there any additional growth that we would have seen in the U.S. has been offset by that captive movement. So it's not just that nothing happened. Two things did happen there. And then in terms of rate expectation, I think that I would suggest that between half year and full year, our rates did increase, primarily driven by cyber. So that will have changed our thoughts around the profitability of the cyber business in particular that we were writing, which would have an impact on the capital both -- sorry, for 2021 and also how we're feeling about rates going into 2022. I'm hoping you're hearing this because I'm getting a lot of background noise.

Yes.

And then I'll go to expenses. So I think the question was, what do we expect going forward in terms of expenses? I pointed out that we -- with our height, we're at about 41% at a more difficult expense point in the cycle. We're at 35% this year-end. I'm not expecting to see any significant improvement in the short term from that 35%. I think our aim should be to keep it relatively flat. And the reason being that we are investing and continuing to invest heavily in things like technology, we've got our digital business, et cetera. And so my AR aim is that -- is to enable ourselves, when we aren't in a position where we're growing by 30% a year, which I think we can all admit we're not expecting to be cross-cycle, we're able to set ourselves up in a position where we can maintain a lower expense ratio and when rates aren't where they are at the moment, the creeping up of the expense ratio doesn't happen than it's done in the past.

Right. Reserves, yes. So we do continue to benefit from the aggregate excess of loss reinsurance that we bought. How much is there left? Enough is the answer. So we're not going to disclose how much we bought or how much is left there, but we are quite satisfied that there is more than sufficient. The loss pick for SCIEX remains slightly above where it was slightly better than it was the year before. And I don't think that reflects a lot of confidence in the business or the fact that there's more to do. I think it's the fact that we are naturally prudent and cautious in our opening aspects. And whilst evidence is coming through that our re-underwriting is working and therefore we're confident to put a little bit more exposure on the books, we still want to open with that, with the prudential margin along with that. So that's just our natural caution coming through, I think. And hopefully, the background noise hasn't interfered too much. Did you get those answers?

I did.

Our next question comes from Andrew Ritchie from Autonomous.

Congratulations, Adrian, on your very strong first year. Couple of quite...

Thank you.

Questions. First of all for Sally. Sally, I remember a discussion with you years ago now where you mentioned there was some negative impact from falling interest rates on your economic solvency position. And is that working on the other way? I mean, I guess I'm just trying to understand what the sort of economic impact of higher interest rates are. Clearly, I can see how the volume yield improvement will come through earnings eventually. But from an economic point of view, is there a gearing? So that's I guess the first question. Second question, at the risk of delving down sort of complicated rabbit hole, in economic terms, you're recognizing the improvement in earnings as you mentioned in Solvency II terms because that's on a best estimate basis, it's coming through quick -- more quickly. If I take that logic though, it would imply that the reserve buffer should be trending up because you're not recognizing intentionally all of the better embedded margin in the business. Now I appreciate there was some usage of the cats margin in the second half. But is that logic not right? Shouldn't the reserve margin be trending up at the minute because there's a big lag between what you're -- what's happening on the best estimate basis versus what you're choosing to recognize? And then the final question, just on capital management. In the past, you've said as you exceed your target level, the first priority would be looking at the debt structure and potentially either refinancing the short-term debt facility you're still using, part of which you're still using or repaying that. But now I think the message is you pay a special dividend first. Is that the change of message there?

Okay. Do you want to do the first one?

Yes, sure. So there would be some positive impacts because we do discount on a Solvency II basis. As interest rates go up, I wouldn't necessarily spend a great deal of time modeling it. It would probably be more apparent for a life company because our liabilities, although we do right liability business and don't have that longer term, so short answer, yes, but not too significant, I would suggest, on the first point. On the second point, you are right. The new one around saying that if benefits come through to the best estimate quicker then the margin should go up, I think that's right. The one thing I would say is that the graph that we show isn't against the best estimate. And there are different assumptions those wonderful actuaries do between the best estimate reserving that they do in the prudential actuarial methodology that we use to show our consistent reserving methods. And one of those is that a best estimate would be quicker to recognize and one with a more prudential margin. But the theory that you show is right, but it's probably more to do with the fact that we compare, it's not a full best estimate within that one. And then on the third, you talked about the debt levels as well. It is something we're conscious of within the debt levels. I think that we will be continuing to look at opportunities. As Adrian said, I'm not sure we're imagining that opportunities for growth are going to go away very quickly anyway. But during that period, we'll be measuring our leverage and ensuring that we're happy with that before we were returning anything to shareholders. But I don't need -- I don't see that happening in the next 12 months because of the opportunities ahead of us. But I don't think I've changed my position, Andrew. I know that conversation. I don't think we've changed our position that the leverage that we have, in particular, the banking facility that we have is something that we utilized in times of significant capital growth, and it's not necessarily something that we have cross-cycle. But we don't have any firm plan in the immediate future to do anything on that yet. But obviously, we'll share that with you if that changes.

So there's no...

Good...

That wasn't intended to be...

That's very helpful. Can I just clarify one comment Adrian made just in the previous. Did you say, Adrian, you expected net growth to be similar in '22 versus -- I wasn't sure we talked about growth on net.

Yes, no, net growth development.

As in net written. Okay.

Yes. Yes.

Our next question comes from Will Hardcastle from UBS.

First of all, just a quick follow-up on that net growth similar to growth. Just how do I think about that if cyber is growing and there was an additional quota share? I just thought the net would be lower. But is there other moving parts elsewhere that changes that? Second question, just thinking about that capital surplus. Just so we're clear going forward, is the 15% to 25% target range pre or post-dividend, just so we can think about moving part in the future, not for now. And then regarding cyber and reserve releases and the prudency point and recognition later, I guess just a quick one, it could just be FX or something like that, 2012 to 15 years on a net ultimate claims basis for the line seems to have deteriorated. It's nothing much, but I was just wondering what the reason for this and why that wouldn't necessarily marry up with those comments.

Okay. So I think in terms of the net and gross written premium, what I said was when I was asked a bit for a bit of clarity about what sort of growth rates to expect, what I said was our net premium growth in 2022 is going to be roughly what it was in 2021. I didn't comment on what the gross number would be. So I wouldn't assume that gross equals net because you're right. You're right there. When we talk about 15% to 25% is pre-dividend and we will look into the 12 to 15 underwriting year as well. The SCIEX, so it's a mixture of executive risk and cyber. So there is some long tail in there. I don't think we thought there was any deterioration, so it could just be FX. But if there's anything other than that in there that's material, we will let you know. But I guess there's nothing to read into it, I don't think.

Yes. At that level of years, it's not a cyber effect going on in that, but we'll clarify the reasons for that. But it's always important to remember that SCIEX -- the act in SCIEX is still a significant part, so I'll take that away for you.

Yes. Okay. That's great. So just coming back to that pre-dividend, 15% to 25%. So effectively, I mean, we've got a scenario analysis typed in, but it looks like the dividend impact is broadly 5 points. Essentially, there may be a year where you'd be willing to take that down to 10% immediately post-dividend. Is that a fair assumption? Because that would -- when we consider that relative to, let's say, potential catastrophe losses, et cetera, that would look very low, I'd have thought.

I don't -- so with -- the 15% to 25% has always been pre-dividend and has seen us good since we brought the regime in. And when -- so I don't see that as a problem. The -- one of the reasons why we've changed the dividend policy that it is a single annual payment and progressive as it is, is so that we have a little bit more flexibility around it so we can make sure it's appropriate to our surplus capital. And with a -- and what other things that we've done, when we looked at where to reset it, it was we compared what the dividends could be to what our expected profits were and the implications that would have for our surplus capital. And I think that the level we set at that should be very affordable for us.

Our next question comes from Ivan Bokhmat from Barclays.

Three small questions, please. The first one is just about the reinsurance. Maybe you could just update us on the excessive loss limit that you have, not necessarily referring to that aggregate program, but what has changed in terms of your purchasing? And you've also communicated that you've spent 24% of premium on outgoing retro in general sense, where should that be going for next year? And the second question is also a bit of a clarification on the admin costs. I understand that the 35% includes both acquisition and your admin expenses. Are you seeing any pressure on your acquisition costs potentially? And what does the mix towards writing more cyber, for example, does for that? And the third question is just on reserving. I'm just wondering, this current 6.4% buffer, I mean we're still in a hardening market environment. I'm just wondering if we should think about this number returning towards the top end, towards 10%, would that be the appropriate thing to do? Or even if that's the right question to ask in the context of IFRS 17. Maybe to put it in a different way, does the 90% combined ratio imply some, is there a buffer addition?

Implies some, sorry, can you...

A change in...

In reserving cost fee, okay. All right. Well, let's start with that one first. So we -- so no...

No.

It does not. And I think one of the reasons why we keep that graph up there is to show that we can produce a decent combined ratio and maintain the same prudence in reserves. Again, as Sally said a few minutes ago, this is not a 6.4% above the best estimate. This is a 6.4% above an actuarial best estimate that we have, which in and of itself has some prudential margin. As a percentage of our best estimate, it's considerably higher than that. IFRS 17 will change those disclosures so that we -- and we will have to make sure that our reserving policy is fit for purpose for that one.

Yes. So we'll be -- the way that we talk about reserves will change under IFRS 17, and we'll be talking about that a great deal in the coming months to take you through that. But our guidance for '22 doesn't take any of that into account. So that's in a like-for-like basis. It doesn't assume that we're doing anything different from a reserving perspective or that the margin would move either north or south significantly. We haven't taken a view on that in order to come out with our guidance of around 90%.

With regards to acquisition costs, we have been working quite hard on making sure that our acquisition costs remain appropriate. You'll note that they have been coming down these last couple of years. It's not really a business mix issue. It's more the fact that we've been, particularly as prices have been rising, making sure those acquisition costs will remain appropriate. One of the things that we've been very vocal about these last few years, and we've been working well with all our partners is to figure out how we can take frictional costs of placing and administering insurance down through improving our systems and processes. So it costs less so that we can maintain our long-term ambition for those costs come down because there's just less expensive work to do, and we'll continue to work on that. So our business mix is not being influenced by what we want our acquisition cost to be. Those are 2 slightly different things. Again, on the reinsurance side, no change to our strategy. So the way that we structure our reinsurance and the sort of limits that we buy across various books of our business haven't really changed. So we buy in the same way to the same sorts of levels that we always have done influenced by how much risk we're taking gross and how much aggregate we have. So there's nothing really to say around any changes to the sort of the amount of reinsurance we buy other than what we've already talked about, which is that as in certain quite capital-intensive lines, but there's a lot of growth, we'll buy proportional insurance to help us do that.

Our next question comes from Nick Johnson from Numis.

Two questions, please. Firstly, on specialty lines. It looks like the growth in that segment decelerated materially in Q4. Just wondering what the moving parts are within that and whether the Q4 growth rate is indicative of the sort of run rate outlook for 2022 in specialty lines? Secondly, on D&O, so the share price collapsed in some parts of the U.S. market so far this year, presumably raises the risk of class action. Just wondered how far outside normal expectations is the current environment around that? And how is the D&O book positioned in this environment? I think you mentioned that there was exposure growth D&O during 2021, but I might have got that a bit wrong.

Okay. So SL growth slowed down in Q4, yes, it did a bit. We've been growing the -- so that -- I think as Bob was saying, there's been relatively consistent growth in specialty lines for a long time actually because we tend to invest in areas where there's natural demand growth, which is why you find us in things like health care and technology and environmental and so on and so forth because we're in those specialist areas where the pool is always growing and it's a good long-term prospects for us. One of the things we've been capturing within specialty lines is a real opportunity in international financial lines, so FI and D&O. When that market dislocated in 2019, I think that growth spurt as we moved to more of a Tier 1 position is reaching its peak now and that's sort of driving the reasons for that slowdown. We're delighted with our position in international financial lines, but I don't think it will continue to grow at the same rate that it has since 2019. Good question on D&O. In the U.S. we have been saying for a while as in across our floor business that we are in a high-risk environment, and that's how we're underwriting. It's what we're pricing for. And so our underwriting and our pricing contemplates a very active, both SEC and other sorts of D&O claims activity. And so we wouldn't be surprised to see that pick up following share price falls this year, and we've underwritten to it, Nick. So our results should contemplate that. Fair?

Very fair.

Our next question is from Iain Pearce from Credit Suisse.

The first one was just on the review of how you look at diversification across the business line. You sort of said you are reassessing how you look at that, the Q3. What's been the outcome of that reassessment and how are you looking at diversification? And following on from that, is there any limit to exposure growth, particularly in cyber based on sort of diversification ambitions of the group? And then secondly, just on cyber. You said previously you've been cautious in recognizing the positive claims trends you've seen since October 2020. Have you recognized any of those in the loss picks that you've made for the 2021 underwriting year? And if you haven't, sort of when should we be expecting the potential recognition of that profitability improvement?

Okay. So yes, I mean, we have been reviewing how we look at diversification. I think what we've done in the short term is to rebase everything back to 2020 rates and look at it in terms of premiums then. And in that way, we were comfortable to grow our cyber premium, which is what we'll be doing this year. You can see from the slide there that exposures are well down from where they were at the peak in 2020. So there's lots of room for us to grow our exposure without breaching the sort of diversification that we had back then. And of course, noting that we're a bigger business now than we were in 2019 anyway, so there's lots of room for us to grow our cyber policy count without becoming less diversified than we used to be. I think we continue to think about how best to model diversification. And as our thoughts progress, we'll keep you up to speed, but we thought the most pragmatic thing for now was to look at it back in terms of 2020 rates and 2020 policy count. Is there a limit to cyber growth? Of course, there is. We want to make sure we manage a diversified business in a way that we always have because I think that's one of the secrets of our long-term success. But as you can see from that slide, we have lots of headroom before we get anywhere near there. So it's certainly not an issue for 2022. And have we begun to recognize some of the improvements that we've seen in our 2021 loss picks? Yes, but only a little bit. And we should be able to recognize more in the next couple of weeks.

Our next question is from Tryfonas Spyrou from Berenberg.

Congratulations on a strong set of results. Just one question on cyber. The combined ratio for the CYX segment is around 93%. Obviously, we don't know the split of cyber and D&O. But when looking at the cyber profitability, do you think -- obviously including the margin you still haven't earned, do you think that is not adequate when used against the volatility in the increased capital requirements of the class? Or do you feel there's still room for improvement there? And I guess how is that reflected on your cautious plans for growing exposure this year?

Yes. So we are looking to put a little bit more exposure on the books as at now. We wouldn't be willing to do that if we didn't think the margin was right and the return on capital was right. And then we had the -- and we weren't able to do that prudently. And the degree to which we do decide eventually to put exposure on the books will be a combination of continued proof that our ecosystem is enabling us to select and work with the right insureds and that we can continue to charge the right amount of money. It's quite a dynamic plan, this one, not only in cyber, but generally speaking. And we will respond and change accordingly to what we see both on the demand side and how well our underwriting and risk selection is working. But we get increasingly confident about that as each month and quarter goes by, noting of course that threats change and evolve the whole time. But one of the signals we want to send today about willing to put a little bit more exposure on the books is that we're comfortable with the margin we think this portfolio has now.

Our next question is from Abid Hussain from Shore Capital.

I think I've got 2 very simple follow-ups. You may have actually answered them before, but apologies if you have just with the various connection issues, I'm not sure if I picked up on the answers. So the first one on the investment portfolio and the rising bond yields. I just wanted -- just in very simple terms, I would have thought rising yields is positive for you just simply because you're reinvesting at the prevailing rates, i.e., at higher rates and so your income level should be higher going forward. I know there's a capital loss, but is that more just a mark-to-market capital loss or are you actually incurring trading losses? So I suppose that must really do not hold our bonds to maturity. So just a little bit more about the actual economics as opposed to the accounting treatment would be helpful on that side, please. And then just coming back secondly on the core guidance of 90% or around 90%. To my mind, and I think, again, this was asked earlier. It feels that it should be 90% or potentially better just simply because if you do end up having an average cat year with the positive rates that you've been earning over the last year or so and continuing to do so this year earning through i.e., similar expense levels you're suggesting, no change in reserving philosophy unless the missing PYD will be slightly lower, but it feels like the core ought to be 90% or better? Any thoughts on that welcome, please.

So on the first one, I couldn't have put it better myself that the economics that you described is exactly right. The fact is, is that because we mark to market that were we to calculate our investor return as of today, the movement in the yields would have a negative impact year-to-date on that. But you're completely right that I think we said it earlier that generally speaking, overall, that's positive for us because that ever-growing large portfolio of fixed income is yielding higher going forward. It's been very low for a couple of years now. So you're completely right about that. Everything you said, I agree with. And then on the combined ratio, I think we -- I think it was Freya that asked it earlier, but I'll say it again. So I think that we started 12 months ago out with a low 90%s guidance and then the year happened. And I think we're starting out this year around 90%, which I see is one notch better than low 90%s. So I do see this is an improving picture compared to where we started out a year ago. Your hypothesis of if you're there and you're still getting improving rates, is there a potential to deliver further in the future? All else being equal, yes, because if we're at around 90%s and we're still getting positive rate unless we see a change in our expectations on claims and there is a potential in the future to see that positivity come through, but we are very cautious about guiding to that more specifically because 12 months is a long time in our sector. And so that math would work. But I think we have to see what happens during this year before we can make -- be more definitive on that.

You pointed out a number of tailwinds, which are all true. And we wanted to signal those tailwinds, as Sally said, by producing some guidance that was one notch better than we produced a year ago, which feels like the appropriate thing to do. But all your comments are true.

I hadn't appreciated the one notch improvement in the guidance, that makes sense.

Our next question is from Darius Satkauskas from KBW.

Just one. You might have answered that, but can you provide some sort of color on what catastrophe losses as a percentage of premiums you would expect in a normal year given what you've got on the books right now? Any color would be helpful.

We don't disclose that.

We don't disclose that. I would say that the way that I would talk about it would be that in 2021, we had overall slightly higher than average, but we haven't done that disclosure. So we can't give that detail apart from to say that I would say 2021 was slight -- was somewhat higher than average.

And we'd also point out that -- yes, what we're trying to explain was that the amount of cat risk that we're taking, relatively speaking, has been going down quite steadily in these last few years. It's flattened out a little bit these last 12 months because of what's happened to prices, but it is still a lot lower than it used to be, which I think is the right place to be whilst we finish building our tools to be able to have a forward-looking view of risk for climate change.

And my second question, just to clarify your comment, you said you expect net premium growth to be similar to last year. So we're talking about 20%, right, and slightly higher at the gross level?

It is.

Yes. I would -- yes, I think in the past, we've talked about mid-teens in 2020. We haven't really updated that one. So yes, written and earned on net was slightly different. So it's of around that order net.

Our next question comes from Barrie Cornes from Panmure Gordon.

I've just got a couple. First of all, looking at reserve releases, it looks as if there's been quite a large reserve release in the marine area. I just wondered what's driven that. Looking at triangulations, it could be some in 2019. And the second question I had was in respect to the expense ratio in property seems to have gone up quite a bit to 44%. You have grown the book there. I just wondered if that's likely to continue that sort of level or if I can pull back again.

Yes, you're right to point out the reserve release in marine. They've had a very good year and that marine book, including of assets under investment at the moment. And so we do get quite a tailwind from a rising yield. And inflation isn't really a problem for insurers as long as we are -- as long as we've taken into account in our loss picks and our pricing tools and the market allows us to price for it. And what causes problems is unexpected spikes in inflation that we haven't priced for or considered, and we haven't included in our loss picks. And where we were back in 2016, '17, when we started to talk about social inflation was we had just begun to realize something that we hadn't seen before and the market wasn't really allowing us to price growth. That's a more difficult place for us to be because there's risk that we've taken on that we didn't realize. And then once we did realize it, we weren't able to charge any more money for it. Where we are now, there's a reasonable consensus that inflation is here, both on the RPI side and on the wage side as discussed earlier. And on social inflation, we have taken those effects into account in our loss picks. We've built them into our pricing tools and the market is allowing us to price for them now, which is a very, very different place to be. So if inflation proceeds as we have assumed, it shouldn't really affect us too much. What will affect us is unexpected spikes that we haven't taken into account or are unable to price for which I don't think is the case at the moment.

We have a question from Ashik Musaddi from Morgan Stanley.

Just a couple of questions I have is, you mentioned that the rate renewal rates are up about 24% on the renewal book. Is it possible to get that number XIX? I mean the reason I'm asking is, I mean, clearly, SCIEX has a reason we understand like why the rates have gone up, and it has gone up a lot, like 49% is a lot. So -- but just trying to get a bit more sense as to how do we think about rate renewal on the order book basically? I mean I saw the individuals as well like 10% on property book, et cetera, 13% on specialty books. So all these looks great. But at the same time, I see that you're reducing cat exposure. I think you were saying that specialty lines might be growing a bit less as well this year compared to last year, which I was a bit surprised like 13% prices went up last year. And I mean everyone is still saying that the specialty lines pricing is still going up. So what's the hurdle? So it's kind of 2 questions, rates, XIX and then why the other parts of the business like property and specialty, you are cutting back a bit. That's one. The second is -- and one more question, sorry. Combined ratio improvement like, say, low 90%s to 90% or around 90%s. I mean there is certainly a 2, 3 points of improvement -- implied improvement in the guidance, I would say. But would you say it's solely driven by attritional or would you say it's because of like either extra reserve releases or either lower cat losses because of your reduced CapEx for you?

Okay. Well, I'm going to forget the last question, so I'm going to answer it first. So let's go there. So I think the reduction in combined ratio guidance is a mixture of all those things, right? So it's a combination of what we expect to come through from the prior years and continued impact of rate on forward-looking attritional loss ratios and the fact that we've been growing and the disclosures we've already made around cat. So it's a mixture of all those things that's driving that 2- to 3-point reduction in guidance. I think we've put up on the slide now, the rate change by division possibly.

Yes.

Depending...

I can't do it in my head.

It's there. It's...

Yes, yes.

Okay, yes. So that's good. So that's shown there. You can see it is quite mixed and which is why it's quite a dynamic plan looking forward. Talking about growth, I'm glad you asked the question. Specialty lines has been growing year in, year out at a reasonable lick despite changes in market. And the reason for that is because fundamentally we've invested in markets that are growing year-over-year. So we're not expecting specialty lines to continue to grow or to cut back. The only comment I was making was that there is sort of -- bit of supercharging we've had in specialty lines has been driven by an opportunity we've taken in international financial lines. And I think we've largely taken that opportunity now. The growth rates are going to revert back a little bit, but there are still good long-term opportunities for growth in specialty lines. And you're right, the pricing environment is better than it was. There is some social inflation there. So we have to be careful, but the pricing environment is better than it was. What we show on the property side is not that we're reducing premium, although we're stopping growing. It's just we've been very careful putting exposure on the books whilst we get our arms around climate change and pricing for that book appropriately. What we were trying to demonstrate this morning is that we are very busily investing and building those tools because we're quite enthusiastic to be able to have a forward-looking view of risks so that we can be confident in our risk selection and pricing and that we're pricing for all the payrolls appropriately so that we can grow our property exposures. We think there's a real opportunity for us there once we've got that done and we want to get that done quickly.

We will take our final question, which is a follow-up from Derald Goh from RBC.

Don't mean to drag the call out, but I thought it would remiss not to ask Bob a question. I appreciate it's early days, but having been another cyber insurer previously, just keen to hear your thoughts about any initial impressions about around these cyber proposition? So for us anything you can say around the U.S. strategy and underwriting culture and things like that.

Well, I think in terms of their cyber culture, I've been very impressed with how they're using third-party data to kind of understand the vulnerabilities or lack thereof of their clients. So they're using them to really help identify the best clients, and that's how you've seen the reduction in frequency. In terms of the U.S., in terms of cyber or generally?

Generally, whatever you want.

I guess a bit of both if that's okay.

Yes. No, in cyber, I think in terms of the SME market, we've moved to Beazley Digital, which I think is a great way to have people focus completely on the digital platform. So I like that approach. I think that makes you more focused on reaching out to your customers in the digital environment. And it's other products as well, but cyber is a big -- the big one on that. So the whole team is focused on getting that right. In the middle market and the larger clients, we continue to grow, and we're using the tools that we have for cyber to achieve that. And we are making sure that Beazley Digital and cyber are working together. And then in terms of other products, I think they've -- it's a similar approach. What attracted me to Beazley to begin with was their strong underwriting reputation. And you can see in how they approach new ways and looking for new data to make the best risk. We're doing stuff on D&O in terms of, call it a leadership score to figure out those things, third-party data that can help us understand the quality of our clients better. And this has been pretty impressive from my point of view. So that's what stands out to me.

Thank you, everyone. This concludes the Q&A session. So I'll hand back over to the management team for any closing remarks.

I think that we've prepared ourselves for. And as Sally said, I think -- hopefully, we can do this face-to-face next time. There are fewer technical risks. It was pleasing to be able to produce a result like this, a return to some decent profitability, return to dividend, which we've been trying to commit to and relatively optimistic about the future. And I hope those are the messages that we've managed to convey. Thank you again for calling in, and we'll see you next time.

Thanks, everyone.