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

Good morning, all, and welcome to the Beazley results for the year ended 31st of December 2020. My name is Adam, and I'll be the operator for this call. [Operator Instructions] I will now hand you over to Andrew Horton to begin. So Andrew, please go ahead.

Thank you very much, and welcome, everybody, through our results presentation. If we just move briefly quickly through the disclaimer. We hit the contents of what we're going to present this morning. So I'm going to give you an overview. Then I'm going to talk about the very important topic of ESG. I'll start on the first-party COVID-19 claims and over to Adrian on the third-party COVID-19 claims and over to Sally on the financials, going through the usual elements of financials, hand back to Adrian on the underwriting, and I will come back for the outlook, and then we'll open to Q&A. So if I move on to what's actually happened during the year, of course, I don't need to point out to everybody, it's a pretty tough year. When we started 2020, we thought it was going to be a very exciting year with good growth prospects. Rates were going up. There was definitely momentum in the insurance world. Yields were down a bit on our investment portfolio, but we thought it would be a pretty good return. Of course, a lot of that changed in February and March last year. We started seeing COVID-19 claims coming in, and we'll talk about those. Went into a recession. So we have to start implementing our recession plan, but we still saw good growth. And in fact, the COVID-19 claims, and I think the recession gave more momentum to growth than we had originally thought and a better rating environment than we thought at the beginning of the year. We obviously did the capital raise in May. That was to take advantage of the growth, the increasing growth prospects of 2020 and 2021. And we saw reserve releases, prior reserve releases. We expect them to come back to a more normal level, start increasing from the very low level of 2019. So if we briefly look at the numbers on this slide, we reported a loss before tax of $50.4 million. That was driven by the combined ratio of 109%, which we flagged when we did our quarterly IMS in November. So good top line growth of 19% and rates on renewal portfolio of 15%. Now the underlying exposure growth is greater than the difference between the 19% and 15% because some lines of business are employment practice liability, some of our professions lines we took recessionary action on and held growth in those lines back. And also, we've withdrawn from a couple of smaller lines of business in 2019. So the underlying growth was greater than the 4-point difference between the gross premium increase and the rate increase. The mentioned prior year reserve release is coming back from a very low level in 2019. Sally will take us through that in a minute. Investment income, good investment income at 3%. We took a relatively cautious approach. Again, Sandy will go through that. Capital -- after the capital raise, this ended up in our preferred range of 15% to 25%, were very similar to 2019 at 23%. And we are not paying a dividend, second interim dividend for 2020 due to the losses. But we have this commitment to getting back to paying dividends at the half year, assuming some level of normal profitability. And finally, we've got no change to our COVID-19 estimates, which we'll look at in a second. So if you look at some of the things that actually happened on the business front, it's great to see growth across all platforms. So not only the U.S., which has been growing very well over a number of years, but here in London, good growth across our Lloyd's platform and good growth internationally. We've been investing in our international platform over a number of years. And we've seen good growth in Europe. So our businesses in Spain, Germany and France, as well as the regional business in the U.K., Canada, Miami, for our Latin American business, and continued investment in Singapore into Asia. We've had great feedback from brokers on our claims paying. And of course, we've had a lot of claims to pay in 2020, good feedback on that, which should stand us in good stead of growing our business further in 2021. Innovation is at the heart of everything we do, and we continue to innovate in this virtual world. We were concerned initially it would be harder to innovate in a virtual world, but we've managed to launch some new products and add new products and recruit new teams during the year, and we flagged 3 on this slide. The transmission failure, so for virtual events. There weren't that many live events taking place, but they can still have problems. Product recall, it's a line of business we recruited a new team in 2020. And virtual care, the combination of medical malpractice and cyber issues, if you're giving health advice online. Of course, that took off during 2020. So we grew that line of business in countries we had launched it and launched in other countries. We've now also got an incubation underwriting team where we can actually test new products safely, which is good, under the purview of Adrian. And Adrian is going to talk further about ransomware, but we did see the severity of ransomware claims increased in 2020. And I think we responded as quickly as we could to that increase, and Adrian will take us through that in a second. So if we look at the charts, the only chart I'm going to look at is the top left because I covered all of the others. You can see how the rate of growth has increased as the rating environment improved. So we did start to see some rate increases in 2018 on some of our lines, further in 2019, and a lot more in 2020. Not surprising, we're taking advantage of the rates increasing. And there is going to be still that rate momentum in 2021. So we're expecting a similar-ish rate of growth in 2021 over 2020. We are great believers in growing net asset value per share as a driver of the share price. And of course, we haven't grown net asset value per share in 2020 because we've reported a loss. Not surprising, the share price has responded to that as well as the uncertainties in an uncertain world. We have a long-term incentive plan, which 120 of us are rewarded with, continues to focus on 3- and 5-year growth in NAV per share. We think it's the most important driver in terms of driving the share price, so we continue to focus on that. If I have a look at the things we've been doing on the very important topic of ESG. We published our responsible business report for the second year running, which has more details, so you can actually look in that -- look at that later. We hired a Sustainability Officer, Chris Illman, in June 2020. And he is giving us a focal point, everything to do with all topics around ESG. And ESG is a much broader topic than just purely looking at climate change. But he is helping us think through the opportunities and threats that are available in all aspects of ESG. Gender targets. Gender is really important to us, so we have managed to achieve our target of 35% female in our senior leadership. We've now reset the target at 45% going forwards. And we're now setting targets for race and ethnicity. We've been very, very good at implementing that in a very rigorous manner. We developed a mental health network, which was not well developed, but that's really important in this environment. We've been focusing on our charity partners, who have really struggled in this environment. And we've donated over $100,000 to Renewable World, which is our main charity partner, but we've also supported a number of charities. And our people have donated a number -- have donated quite a lot of time during 2020 to a number of community initiatives. This topic is becoming more and more important to our stakeholders, and therefore, we'll be listening more to them. And our aim is in the first quarter to publish our sustainability strategy. So we'll be publishing that in March, along with some diversity targets. I move on to COVID-19. As I mentioned at the beginning, we haven't changed our numbers for COVID-19. So I think it was in September last year, we announced we have $340 million of claims for our first-party business. The majority of these claims are related to our event cancellation, our contingency business, but it also include property, mainly business interruption in the U.S., and reinsurance and a small amount of marine. So we're still comfortable with that $340 million. We also said there will be $50 million of further claims if the world doesn't improve in the second half of this year because events will be further canceled, and we're working with our insurers on that. And they are comfortable and hope that events may take place in the second half, but we still have that potential downside, and it hasn't changed. The good news is claims have been -- being settled in line with our estimated provisions, and that's why we feel comfortable with these estimates. And we have been looking at action to re-underwrite COVID exposed lines, which Adrian will talk about in a second. The final comment for me is we did successfully ship -- shift to operating remotely, which was great. So we didn't miss a beat when we switched into operating remotely, and almost everybody is working from home at this point in time. And we continue to focus on the well-being of all of our people. That has been a high priority throughout all of this. And it's probably even more challenging in these colder, darker months of January and February, but we continue to focus on that. I'll now hand over to Adrian, who will take us through the COVID-19 third-party claims.

Andrew, good morning. I'm Adrian Cox, the Chief Underwriting Officer for Beazley. So as some of you may remember, this time last year, we did quite a lot of work thinking through the sorts of liability claims and exposures we could expect from COVID and the associated downstream impacts like recession, and addressing those in our underwriting through risk selection, price coverage, terms and conditions, and so on and so forth. And we relaunched those new underwriting guidelines for the April 1 renewals. Our central expectation back then was COVID liability claims will begin to manifest as we emerge from the pandemic and looked back and started to judge who did well and who did less well. Originally, we thought that would be the end of last year. This whole thing has taken longer than I think anyone had hoped. And so we're now expecting those claims to manifest sometime in 2021 in this year. You can see from the chart on the top right here that although there have been some COVID and related claims, they are, as yet, quite limited. They were, in fact, more in the first half of the year than the second, and that reflects insureds looking for first party, particularly business interruption coverage wherever they could get it. The pink lines there show not only COVID claims, but also recession claims and racial discrimination, because we're tracking all those sort of mini cats. Having said that, although there haven't been many, from the research that we've been doing about what sort of claims are around and our own experience, we do have a picture emerging of what we can start to expect. And I've given 3 examples there. On D&O, there are about a couple of dozen FCAs securities class actions so far around COVID alleging, broadly speaking, misrepresentations. If you're a vaccine company, you may have misrepresented how far advanced your research was. If you're a cruise company, you may have misrepresented how clean your ships were and so on and so forth. Or you may have misrepresented what you're going to do with the payment protection plan loans if you took them from the U.S. government, that sort of thing. On the health care side, we have seen some claims around failure to prevent or transmit the virus and from a rationing of health care and environmental around cleaning up pollution. So the good news is, I think if there is good news about this stuff, that those are the sorts of claims that are -- that we had contemplated when we put together our underwriting guidelines this time last year. So we haven't had to pivot to counter anything unexpected, either by line of business or by types of allegation. As I said, we do expect this to start to gather pace this year as the economies reopen, and we will absolutely adjust as we learn. And again, I think the good news is that as we speak, 10/12 of our book has been written new or renewed with these revised underwriting guidelines. So in a funny sort of way, a longer pandemic has helped us make the liability business more COVID resilient, remembering, of course, that the vast bulk of what we do is on a claims-made basis, so that gives us that ability to pivot accordingly. And with that, I'll pass over to Sally to take us through the financials.

Thanks very much, Adrian. Good morning, everyone. My name is Sally Lake, and I'm the Beazley Finance Director. I want to take you through 3 areas before passing back to Adrian to tell you more about the business. I will cover investments, reserves, and finally, capital. But just before we go into that, let's firstly revisit the financials briefly. Andrew has covered most of the main numbers already, but I just wanted to touch on the expense ratio improvement. And there's a few reasons why we're seeing this, some are one-offs and some are more sustainable. And the main 3 I bring to your attention is that, during a period of significant growth, which we're definitely in from what you've seen, we do actively aim to improve our expense ratio because now is the time to do that. Secondly, clearly, the financial result wasn't where we expected it to be at the beginning of the year. So the remuneration that we allowed for at the beginning of the year has obviously reduced in that time, which reduces the expense ratio as well. And finally, and not as big in numbers, but we do spend money on traveling and entertaining people. And while it isn't a large amount of money and we aim to manage it actively in all years, clearly, that reduced dramatically during 2020. So like I said, there's a mix of some things that are one-offs and sustainable, but we continue to actively manage our expenses. Moving on to investments. And I think that we're very pleased with the overall return that we achieved during 2020. The conditions were extremely volatile for much of the period. And the investment team did really well in taking swift action to manage our risks in these unusual circumstances. Coming back to Q4, and we continued to see positive returns, especially in capital growth assets, and we ended the year at 3% overall. If I now go on to the portfolio. Firstly, I'd point out the balances that we have now. We've gone through $6 billion in terms of how much money we have to invest, clearly, because we've been growing significantly of late. So that's an interesting thing to think about, because even though we're expecting lower returns in terms of percentages, the amount of which we're going to achieve via that percentage has increased significantly over recent years with the growth in the business. The split between our core fixed income portfolio and our capital growth assets has little changed over the year. We tend to stick to an 85%-15% split. Clearly, during the year, that changed at certain points, as I mentioned. But year-on-year, we're broadly similar. The thing that's worth noting is we have made a switch to more sovereign debt within our fixed income portfolio, as the very low credit spreads currently available in investment grade offer very little return for the additional risk, and that's our thinking at the moment. Obviously, we manage that on an ongoing basis. Other asset classes, the exposures changed little year-on-year, but they varied again significantly throughout the year depending on the conditions. Worthy of note that our hedge fund portfolio performed very well during 2020, with a return of nearly 9%. We remain cautious while looking at our risk, and we activate results on an ongoing basis. Most recently, we added risk as market conditions have been improving, but we focused on entering positions that we can easily reverse out of if some volatility just come back in. Unsurprisingly, the overall yield on our core portfolio has reduced, and at the end of 2020 was at just 0.6%. But as I said earlier, we have that being applied to a much larger balance in the last few years. Now on to claims releases. So what we're really seeing on this graph is the benefit of the re-underwriting that has been going on for a couple of years now within some of our shorter tail lines. Marine, property and treaty, who also reserve strengthening during 2019, have moved into reserve releases this year, and that's really pleasing to see. It's also very pleasing to know that they're continuing to receive changes in rates as well, so hopefully, that momentum will continue. Given the recent challenges in the CyEx book, which Adrian will discuss more later, we saw a small overall strengthening in this class. And finally, our largest division, specialty lines, saw good a experience on mature years of business and produced an increase in terms of releases compared to the previous year. So that's looking back. So we'll move on to our favorite graph. So what does the future hold? And what this graph aims to do is give an idea about what's in the tank for future reserve releases and to remind everyone what this graph is doing, it's comparing the reserves in our balance sheet to a bottom-up actuarial estimate, which in itself has a level of prudence already embedded and is not a best estimate measure. Now what we do is we aim to hold within 5% to 10% range of this graph. And this year, we ended about 6.3%. Now this varies year-on-year depending on the claims experience for that year. And in particular, this can be affected by -- in years where you have significant specific losses. And clearly, in 2020, we have that situation. There's actually 3 things I'd like to point out that are affecting this graph and making it lower than it would be otherwise. So firstly, and we've already mentioned around the COVID losses of $340 million, they held at a similar amount between the actuarial estimate and the -- on the balance sheet. So there's less margin there would be on that reserve compared to our general reserve. Secondly, we've had some natural catastrophes in the second half of the year, lower than in previous years, like 2017, but still adding to this effect somewhat. And finally, we spoke at the half year about the reinsurance that we have on our specialty lines and cyber CyEx books. And this aggregate excess of loss reinsurance is actually currently recovering under a number of years within those books. And as we look at the surplus after all reinsurances apply, after that's being applied, you'll have similar levels across those measures, even if they're different before that reinsurance is applied. Were it not for these effects, we estimate that this graph would have landed around the middle of this range. And moving on to -- finally to capital before I hand over to Adrian. So before we speak about the numbers and where we ended the year, I wanted to give some context for how we think about capital at Beazley, as we do it in a way that is different than other approaches that you may be more familiar with. We focus primarily on the Lloyd's requirement, which, whilst based around Solvency II, differs significantly from more commonly used Solvency II solvency approaches. So firstly, whilst most capital calculations are based on a 12-month time horizon, a 1-year basis, the Lloyd's requirement is actually based on a -- on what capital you'd need to an ultimate position, which clearly makes this number higher than a 1-year. This varies year-on-year for us. But generally speaking, we think this adds around 20% to 30% to our number. Secondly, Lloyd's apply an uplift to that ultimate number before they come up with their requirement. That number is currently 35%, and that is already allowed for in the numbers that we quote. We then target to hold 15% to 25% above that number. And so if you compound those various effects, moving from a 1-year normal Solvency II basis, ending up in a place of 15% to 25%, will be comparable to a solvency ratio in the high 100s or low 200s. So within that context, where did we end the year? So as Andrew has already said, we ended the year within our target range of 23%. And to remind everyone, this is after allowing for all the business that we plan to write during 2021. So we've got our business plan. We're executing on it. And all of the capital required for that has already been baked into these numbers. Additionally, it's worth mentioning that our U.S. admitted carry has a great success over the last few years and has been growing significantly. This has led to the capital requirement which is based on a formulae -- risk-based -- local risk-based capital approach, rising significantly in recent years. And we expect it to continue growing, as we've already said, in terms of the future. Then what we have done at the end of 2020 is we've set up a captive arrangement in order to manage these capital requirements as they increase. And so if you recognize the number that we're showing for the U.S., it's slightly lower than we expected to be at the half year because we've introduced this. Finally, Adrian is about to speak to you on various things, but he'll also speak -- within that, he'll talk about some contingent reinsurance that we've bought, which doesn't impact the numbers here. But it's worth noting that, that gives us an additional lever during this exciting period of growth should we need it. And with that said, I'll pass over to Adrian to talk about the business.

Sally? Okay. So as we enter 2020, as Andrew mentioned at the beginning, we were sort of moderately optimistic that the market was moving in the right way, having really opened up and confessed in 2019 about the profitability issues across a number of different areas. And that market sentiment moved materially in Q2, I think, to address much more urgently those existing underperformance issues, and notably, social inflation, and the new issues around COVID-19, recession, racial discrimination, which emerged soon after, and cyber, which I'll talk about in a little while. And that can all be seen on the graph on the right-hand side of the page. We started the year at about 6%, finished at 15%. You can see the pause there in March, April as we went into lockdown. And then the real change of pace in May and June. And that for us led to quite a change in the risk reward across a number of parts of our business, including property, marine, treaty, D&O, for example. And so in response to that, our growth rate really started to accelerate in the second half of the year and particularly the last quarter. And that momentum has really continued into 2021. More so, I'd say on the insurance side and the reinsurance side. But all our teams actually are more positive about the market environment than they have been for a very long time. The impact of new capital is relatively limited as yet. It did take, I think, a little heat out of the reinsurance renewals at 1/1, which has been much commented on, and was actually fairly helpful for us as a buyer. But on the insurance side, I think the brokers have mostly been using new capital to fill gaps. Not many new carriers are actively quoting, and none are being particularly aggressive or disruptive. So for us, I think this represents a year to capitalize on a rare moment in time to grow the business at pace and really to form a foundation of our client base for the next few years across all our platforms, the U.S., internationally and in London, and that's very exciting. It's always better, I think, to grow in a relative vacuum when your clients and your brokers are asking you to take more of a prominent position and to lead more and to step up. It's a much better time to grow than when you're having to force your way into things. We're still able to apply our diligent, stringent underwriting areas, underwriting particularly in areas of elevated risk, which I will talk about in a little bit. And historically, of course, these sorts of situations don't last that long. So whilst the new capital isn't particularly frustrating as now, you can sort of hear the whos in resistance. So we need to take advantage of these conditions while they are with us. You can see from the chart on the bottom right of the page, we have split our business into 4 categories, sort of business as usual, which we're happy to grow, areas where we're actively growing our market share because we think this is an opportunity to do so, closing, continuous monitoring and restraining. The bulk of our business is in those first 2 categories. We have about 10% in the red bucket there. And that's areas where we have some elevated risk where we don't think we're being adequately rewarded for it yet. And again, as Andrew has mentioned, that's in part of our management liability book, particularly employment practices across some of our PI lines and some of the med mal. And then the blue bit there is the -- is our cyber business, which again I'll talk about in a little bit, which we are very closely monitoring. So in helping to ensure that we can maximize this opportunity, we bought, as Sally mentioned, a contingent quota share. And by that, we mean that we've agreed with a small cohort of our close partner reinsurers that we can, actually, when we want to, seed up to $200 million of premium on a quota share basis of our liability business pre-agreed terms, which gives us great flexibility as we go into this year. I've mentioned elevated risk a couple of times, so I thought I'd talk about that. I'll just move on here. So our underwriting contemplates that we're in this period of elevated risk. And the reason we're still able to grow, I think, is that we can underwrite and price for that elevated exposure. You can see on the -- from the chart on the top right, that new liability claims were actually down last year. And I think that's understandable. A huge amount of reduced activity across the world. I sense this was a crisis that we all need to pull together on so not a great time to ligitate, and the courts are closed. But in our view, this is all temporary. The drivers of social inflation and claims inflation we talked about last time haven't gone away so we do expect this to bounce back. And I think a good example of that is, if you go to the bottom right chart, which shows you U.S. D&O securities class actions, and the top red line there shows that, indeed, in 2020, there were fewer than they were in 2019. But a reasonable chunk of those FCAs for the past few years have been around M&A activity. And of course, M&A went into stasis in the middle quarters of last year. So if you correct for that, actually, it's relatively flat. So none of this has gone away. And in addition, there is still a lot of uncertainty about how much of a blame game there will be around COVID-19 and how much of a recession we will have to tackle. So our central expectation is that the claims environment will go back to where it was this time last year. But as I mentioned at the bottom of the slide, we are in this market able to take the underwriting action and we need to through coverage, terms and conditions, pricing and exposure management, and not least, to do the due diligence that we needed to do. And the last area I thought I would mention is cyber, and our view over the last 12 months and how we're addressing the issues that has thrown up. Frequency spiked sharply in 2019 and stayed high last year. The threat has changed, but the overall level remained the same. So we did, as a lot of people have commented, see more attacks that use vulnerabilities of remote access, but the overall numbers of claims stay relatively flat. What happened was an increase in severity, driven by the fact that our clients were paying more ransomwares than they had before. And that was driven by the increasing sophistication of those ransomware attacks and an increasing prevalence of data exfiltration. So not only were the bad actors taking down systems, they were also stealing data and threatening to sell it or release it to the public. And it was that twin attack that really persuaded a lot more people to pay the ransoms and to pay them at increasing prices. We've done a lot of work in 2019 and into 2020 to start to build a capability of assessing live that threat environment and scanning our clients for those vulnerabilities. And we knew we had to do this as malicious attacks started to become more meaningful a few years ago, but it's a capability that takes some time to build. And we rolled this out last summer and adjusted our underwriting accordingly. And one of the first insights we gained was, how many of our clients had these high-frequency vulnerabilities, which were fairly old and for which the cures are either patches were well known? And the conclusion we drew from that was that network security in the corporate world was actually, a, quite variable; and b, not very good. And therefore, the proximate cause of increased ransomware claims isn't necessarily the increasing number of bad actors, it's simply a lack of decent protection. And the analogy that we started to draw was that if you leave your windows and doors open, you shouldn't be surprised if you get burgled. And so the real solution here is setting standards, enforcing standards and good risk management. And this, I think, is a familiar role for insurers. It's something we've done for a long time. We started selling fire standards in the U.K. after the Great Fire of London in 1666, when we started building firewalls between houses. When you ensure your factory against fire risk, you expect your property insurer to send around an engineer to inspect your building and to make recommendations, the same with employers' liability and health and safety. So our strategy here is to begin to perform the same role we do elsewhere, which is to set and enforce network security standards to bring the number of attacks down, as those -- that's the proximate cause. And I think there's going to be a sea change in how cyber insurance is underwritten and risk managed by insurers, particularly in the mid-market where most clients don't know and wouldn't be expected to know what those standards are, particularly as the threat continues to evolve. In addition, of course, we are re-underwriting. Our risk selection has changed a fair amount in the last 7 months as a result of our additional due diligence and the scanning that we're doing. Plus, of course, we're managing limits and deductibles carefully and repricing to take advantage of the increased frequency and severity. And the market, of course, is moving very fast. It's the most dislocated one we have at the moment. The highest rate changes across the book are expected in cyber this year. So we are able to do what we need to do. And that's useful for us as we adjust our pricing, but it's not the long-term answer. Consistent and prudent risk management and engineering solution is the real answer to that, and that's what we'll be rolling out this year. So whilst we do that, of course, we will be closely monitoring and tracking activity and making sure that the activity we see going forward correlates to our re-underwriting, and we'll continue to adjust as we learn. And with that, I'll pass back to Andrew for the outlook.

Thanks for that, Adrian. So let's get the crystal ball out and look at what's going to happen in 2021. Of course, the crystal ball last year was generally pretty accurate, other than one major thing that it missed. So let's look at 2021. We feel we have an exciting outlook. As we've talked about, the engine has grown. So we've been growing the premiums as the rating environment has improved. Sally mentioned the investment balance is that much greater. So we have good growth prospects going into 2021, with continuing rate momentum over many of our lines of business. As Adrian pointed out, there are certain lines we're still taking invasive action, and we continue to monitor cyber very closely. Good capital position to finance further growth as we go into 2021 and 2022. It's always difficult to estimate what's going to happen a year hence until we hit the summer of this year. And finally, there's commitment to get back to paying dividends. So we're very committed to returning to dividend payments at the interim. On the back of that, we are now ready for some questions, so we can go to Q&A.

[Operator Instructions] Our first question today comes from Kamran Hossain from RBC.

The first question is on capital. I guess looking at the numbers, whilst it wasn't as bad as most people had assumed, you still made a second half loss. Can you maybe walk through how you got from, I think it was a 22% ECR surplus at the first half to the 23% now? What quantum of benefit is there from the captive and the reinsurance? And then also, how permanent are both these measures? I assume the captive is probably relatively permanent, the reinsurance perhaps rolls off in a couple of years. The second question is on the combined ratio. You're talking about low 90s for 2021. Firstly, does that include COVID at $50 million that you've talked about? And when will you really start to see the mid-teens rate improvements impacting this ratio? Is that 2022 or beyond?

Do you want to start on the capital one?

Sure. I hope I got some points, so I'll do my best. So what's happened in the second half of the year? So clearly, everyone was aware of the further increase on the COVID losses. But also, depending on what happens in the profit expectations versus actual, that will move it as well. And what do we see? So we saw investments improve a lot compared to what we're expecting. Clearly, we're in the middle of the year last year. We weren't expecting the rallies that we've seen. So that's one thing. And also, it's affected by everything else apart from COVID. And as Adrian and Andrew both alluded to, what -- other areas of the business, every -- the business is doing really well. And so what we've seen there is positive momentum there. So taking all of those into account along with the COVID losses has led us to that position. Just to add on the captive. I think at the half year, we estimated the requirement to be about $300 million at the end of this year, and that's reduced down to -- I think it's too -- I can't look at too many screens around, just under $250 million. So that's been the benefit of actively managing that capital there, which clearly will help as well. So it's a combination of those 2 things that have happened. And to be clear, on the quota share that Adrian spoke about, that has no impacts on these numbers. It's just a worthy note as an additional lever over and above everything lies by -- per our capital. In terms of permanency, so the captive is a -- something that we want to continue using, for the reason that we are continuing to grow our U.S. operation. It's extremely important to us. We like the business there. And so we expect that to continue to grow, and then therefore, we'll continue to use the captive. On the reinsurance, I'll have a go, Adrian. We renew annually, but it's very rare that we do things on a 1-year basis, but it's something that we'll continue to monitor over time. Adrian, I don't know if you want to add anything to that one.

Yes. Some of our reinsurance is on a rolling 2-year basis and some of them are annual. But we don't tend to do things staccato. And then -- and that's one of the reasons why we have a close bunch of reinsurers that work with us for a very long time. So we'll be able to use that, I think, as long as we need to.

I also think we're quite careful about the reinsurance and thoughtful on the reinsurance because there are 2 lines of business where we've added more reinsurance. They were mainly on the liability side, and they are more capital intensive and they can be more volatile. Now the great beauty about that is we can start dropping the reinsurance and increasing our net position as the rest of the organization grows around it, with the growth prospects in those in D&O and once the other lines is so great in 2021, we have decided to hedge some of that out. And I just pick up the -- you had the question about the low 90s, which is a relatively straight one -- straightforward one. No, we have not assumed the $50 millions in the low 90s. Because if we thought the $50 million was not going to -- it was needed, we would have booked it in 2020. So we are at this point in time talking to our insurers, have the belief that $50 million won't be needed. But there is still the possibility that the world that doesn't improve, and therefore, that will be the impact on us. So that would take up the combined ratio, I guess, by 1 or 2 points, Sally. Based on our net earned premiums, it's a relatively straightforward calculation based on our net earned premiums. The mid-teens is always an interesting one growth, because on short tail, that will come through pretty quickly. So if we're in property and marine and getting rate increases, that comes through relatively quickly. And you would see the impact of good rate increases in 2020 potentially coming through our 2021 numbers. The medium tail, as we've had claims inflation over a number of years, we tend to be more prudent and wait for about 3 years. So you'll probably start seeing the rate increases of '19 and '20 coming through 3 years down the track on the medium tail book as opposed to straight away, when we'll have a great idea of where the loss costs are going.

Our next question is from Andrew Ritchie of Autonomous.

Good to see you all. Couple of questions. First of all, you've referenced several times in the reserve disclosure to the benefit from aggregate reinsurance on some of the years, I think it's on cyber and specialty where you've increased loss picks on some prior years. What is the kind of remaining limit on the aggregate? I guess I'm just trying to judge the degree to which gross might become gross as opposed to gross with a big drop-down to net, which it currently is helping you. Secondly, there's a comment that you're reviewing your realistic disaster scenarios. I don't know whether it's a worry about that comment or not. I'm not sure what you're implying by it. It's just mentioned a few times. What do you think has gone wrong? Or what do you think you need to review on that? Third question. Sally, I understand your view of required capital is harsher than a standard Solvency II, but that your review of available capital is also more generous because you don't tier the debt. Maybe just give us a sense. What is your tolerance on the amount of leverage you want to include as available capital? Where are you on that? And is there a thought maybe you should introduce some kind of leverage tolerance level? And the final question. I'm left a bit confused as to what the impact of the new reinsurance you purchased will be in '21. Because it sounds like, obviously, the contingent quota share I can ignore unless you really need to draw it down. So is there -- what's the difference likely between gross and net growth in '21?

Okay. So I'm going to -- just on the realistic disaster scenarios. In very straightforward terms, Andrew, obviously, our realistic disaster scenario for a pandemic was not as extreme as what's actually happened. But not surprisingly, one of the lessons learned from that is we need to review the other RDSs and think through the unthinkable events. And therefore, it seemed to be a natural lesson learned that we would go through the realistic disaster scenarios and other areas. And Adrian has been leading that work to confirm that we think they are good enough realistic disaster scenarios, because not surprising, we have a number, whether it's terrorism, political risk, nat cat and so on, cyber, we have a number. So that's all we're flagging. We just thought it was a natural thing to do having found out that we had more event cancellation losses than our RDS for event cancellation was. Sally, did you want to have a go -- which one have a go? We never got to require -- any of those, you have a go at any of those, and we can come back to the new RI with Adrian at the end.

Yes. Okay. I'll do what I want, and then Adrian will do the rest. That's always -- that's what I apply most days. So on the capital, it's a really good point, Andrew, and I didn't mention that in my voiceover. So just to get everyone up to speed. So at Lloyds, you don't have restriction on the resources that you can use for your capital. So that does -- you're right, that, that does complement -- it could complement things. So in terms of leverage, I think on a medium-term basis, and I'll tell you why I say that, I don't think we'd want to go much more above where we are, possibly a 2:1 ratio would be the thing I would always think about. It's not a target, but that's what we keep in mind when we think about the leverage. It's a medium term because, clearly, we have $225 million of banking facility unutilized, which is available to us extremely quickly and that is available should we need it. And I would be comfortable utilizing that if that was for a short term before we looked at other options. So that's how I think about it. I don't -- do I want to go -- to have another -- do I want another target in my life? Probably not. But hopefully, what I've said gives some comfort around how we think about that. What we're focused on is growing the equity at the moment, and we've got the real -- the market conditions in order to do that. I'm going to start on the stop-loss and then hand over to Adrian. I think we talked at the half year about the amount of limit that's available per year on the aggregate reinsurance. And I think from memory, it was -- and I'm going to -- please check this, but I think it's around $165 million in a year. There is some aggregate over a number of years that we've got a great deal of limits per year. And you can see the effects in the loss development tables of what there -- of where we're getting that recovery quite clearly. The one thing I would note on this is that we're reporting very immature years getting the benefit of the stop-loss. That's very early and has margin within it. So as we evolve that over time, we'll see how that goes. But remember that that's where we held quite a lot of margin in the early -- the immature years as well. And we're recovering at the moment, but then we'll have to see how that margin comes out over time. But Adrian, you're the expert.

I guess, right. Sally, we bought some sort of aggregate protection on the long tail business since 2011, and we've generally bought it in 3-year chunks. And so it's been a consistent part of our strategy. There is plenty of limit left. We buy in percentage terms so the amount of limit that we buy grows as our book grows. And it's -- and the reason we bought it was for exactly this sort of confluence of a number of different things happening at the same time. So -- and as Ally -- as Sally says, we've increased the loss picks in some areas, but we haven't decreased them in areas where we think there might be margin, but it's too early to say yet. So the ultimate position of this is relatively unknown yet on the more recent years. I guess the bit that's left is the gross and net. So you're right, we haven't triggered a contingent credit share yet because we don't need it, and we don't want to buy more than we need. We do want to grow on our net business and take advantage on our own balance sheet with the conditions that we have. So as Andrew mentioned, we bought in the more capital-intensive areas. We get that ultimate business mix. The only thing we bought new this year is a quota share, an additional quota share for the D&O and FI business, which is growing at a very fast rate. So our net book this year will grow slightly less fast than our gross, but not a great deal. I think the only other comment I'd like to make on the RDSs is that the -- you have RDSs, the risks where there's not a great deal of information, and so you put together scenarios because you can't build a stochastic model. But what we decided to do was, instead of having one RDS, we design a number for each scenario. Each one of which is -- has a different level of stress attached to it. Do we get more idea about how things behave if they get really extreme and not rather than just rely on the one number, which the market has done historically? And that is in direct response to the fact that most of the market had a PML for contingency, which proved to be wrong. We might have learned a bit more about that if we stressed them before, and that's what we're doing from now on.

Our next question comes from Emanuele Musio from Morgan Stanley.

I have 2 questions. The first one is following on the Kamran question on capital. In September, you said that your solvency was at the bottom end of your target range. Perhaps you already answered. But I was wondering, how would you split the improvement across all the various contributors that brought the surplus to 23%? Maybe if you don't have a breakdown, how would you rank them? And also, a question on reserves. What kind of assumptions have you made in relation to casualty trends? And I mean do you see material risk potentially emerging from litigation on the back of COVID-19?

Do I start, Andrew?

Yes, sure.

Yes. So we -- what we said at the third quarter was we expect to be within our range at the half year. I think the maths that was generally done at the third quarter was to take $170 million of our half year, which mathematically would have got you from the middle top to the bottom. So -- but we didn't update at that time. And we don't update out of half year because things are always moving. In terms of ranking, I definitely -- I don't -- I can't give you a split because I generally don't have one. I've not got it here that I'm not telling you. I think it's a real combination of the 2 things. I mean you can see -- you could look at the reduction in our expectations on that. On the U.S. admitted, you can do that math because like I said, I think it's around $50 million, which will be a couple of -- would be a couple of points. And then the difference, if you think about the $170 million being offset by captive investment in claims, a couple of points from the captive there. And then the rest mix between investment and claims experience on everything else, I think. On the reserves, and again, I'll do my best, and then Adrian will no doubt have a view. So what have we assumed on casualty trends? We definitely -- whilst we've seen some slowing down, we definitely haven't taken any credit for that. And we're assuming that it's a temporary thing that will come back to life when it's allowed to. And so we haven't taken any credit. We've noted it, but not taking any credit for it as such. And in terms of downside, we obviously have our margin -- going back to what we saw for the half year, we have our margin. We have this aggregate excess of loss insurance as well on the books that take a while for the claims to really show themselves. But again, Adrian, do you want to add?

Yes. I mean we went through our book line by line and applied loadings on loss picks for 3 things, the 3 catastrophes that we were dealing with: COVID-19, recession and social inflation, turbocharged by an expected increase in racial discrimination claims after the tragedy around George Floyd. Those loadings remain in place. As I said in the slide around elevated risk, although we've seen a temporary reduction in liability claims, we expect them to bounce back. And we expect the claims environment to be as difficult or if not a bit more than it was at the end of 2019. And of course, that was the big discussion in 2019, about how we were dealing with it. And our central expectations around that haven't changed. The only thing that has changed is what we can do about it, right? And we're able to price for it and underwrite for it in a way that we weren't as much a year ago.

Our next question is from Iain Pearce of Crédit Suisse.

The first one was just on cyber. There seems to be quite a noticeable step change in the level of rate increases you're getting in Q4. So I'm just wondering if you could talk a little bit about what's going on in that market more from a sort of competitive dynamic standpoint. I mean the sort of rate increases that you're achieving probably implies some line exits and that some people out there really feeling some pain. So I'm just wondering sort of what you're seeing on the competitive standpoint. And then sort of aligned to that, the level of rate increases that you're seeing, how does that compare to your expected loss cost trends because, obviously, there has been quite a step change there? And then just a second quite quick one on the dividend. When you talked about the dividend resumption, is how we should be thinking about that a resumption in line with where it was before it was suspended?

Let me pick up on the dividend. The dividend has got to be dependent on where we actually are at the half year. So let's see where we are at the half year. And then commitment is, it's a great financial discipline and we've always had this financial discipline, as Sally was saying, on the reserve strength. And then the capital surplus, this is sort of the third leg to our financial disciplines. Still, we want to be a dividend-paying company. But we have to see where we are at the half year before we can be definitive of exactly where we're going to be. Adrian, over to you on your view of cyber landscape.

Cyber? Yes. So I mean, I think the sort of context to this is that, for the sort of claims that we've been talking about here, it is a very short-tail class of business. So you can write a risk on Monday. It can get a ransomware attack on Tuesday. And then you parted with the ransomware -- sorry, you parted with the ransom to get the key 7 to 10 days later. And you don't usually have 10 days between when something happens and when the insurance pays out. And so that means it gets noticed very fast by companies because insurers generally hang on to the cash for longer than that. And so it's prompted a much faster and more visceral reaction than it would do for other claims, which take a lot longer to pay out. And that's why it's been so swift and also so varied. And so what we're seeing from the market at the moment is a very wide variety of reactions. It's a dislocated market because it's not an equilibrium. There is no commonality to what insurers are doing on mass. Some have withdrawn. Some have reduced their writings a lot. Others are trying to reduce their limits. Others are trying to impose sublimits or withdraw coverage altogether. And others are trying to shot prices up and anything in between. And then so -- and that all really happened towards the end of Q3 and into Q4 last year and has accelerated sharply. I mentioned that we're expecting the rate increases to be highest in cyber. That's certainly the plan. And the January rate increases that we got were in line with that plan. And the marketplace is still moving. So I think that's reassuring in a way. But as I said before, I think the biggest thing for us is that we want to make sure that the way that we are scanning for vulnerabilities, scanning our clients and risk managing our clients starts to result in the number of ransomware attacks coming down because that's evidence that our re-underwriting and our strategy is working. And that's as important as the rate changes that we're getting.

Our next question is from Darius Satkauskas from KBW.

Another question on capital, please. At the end of your press release, you discussed that the capital surplus is currently $476 million or 23%. I guess I'm interested to hear how you think about your surplus target range. And if we should think of the bottom end of the target range as a hard line, would you be willing to cross that bottom end of your target range if you expected good growth opportunities next year, for instance, as well as wanting to pay the dividend? That's the first question. Second question. You guided to low 90s combined ratio target for 2021. Can you give us some detail how you think about your reserve releases in 2021 to get to this target, particularly as you have higher loss fix and specialty, some liability and cyber lines?