Good morning, everyone. Good to see you all in real life. First time in a while. Thank you for coming to our 2022 interim results presentation. Do forgive us if there's the odd slip up doing this live for the first time in a while. We'll see how it goes. All right, I'll kick off and then we don't have any guest presenters this time, so it's just Sally and I. I will kick off and then pass over to Sally for the financials. To start with the overview then. I'm pleased with the results. You know, an underwriting profit of $232 million for the half year is a record in dollar terms and the best combined ratio we have turned out since 2015.
ROE of only 1%, but that's driven by the $193 million of investment losses, which is mostly mark to market. That record underwriting profit is a combination of good growth, you know, slightly higher than we thought at the beginning of the year, of 26% and even higher net, which Sally will explain later. That combined ratio of 87%. We've achieved strong growth really across all divisions, particularly cyber. The division growing the least is property risks, and that's because of our continued caution towards natural catastrophe exposures as we continue to develop the tools we need really to be able to get that forward-looking risk view that we've been talking about for the last year or so.
The rate change of 18% is slightly higher than we'd expected, again, influenced by cyber. As you'll see shortly, at a total level, all our divisions have received or have achieved rather positive rate change this year. The combined ratio was lower than we'd forecast at the beginning of the year because both losses and expenses are better than we had originally expected. We're expecting expenses to revert back towards the original budget of 35% towards the end of the year. Claims have been lower than expected despite the war and other headwinds, which we'll discuss, and our surplus over actuarial reserves pretty stable. We'll explain the movements there more in more depth later. Our claims surplus essentially hasn't grown because it hasn't not grown because we've released more.
On the contrary, we've made further provisions for inflation and other things. It's because our initial loss pick for 2022 is lower than it has been for a while for an opening loss year. Because of the positive rating environment, partly, but also because we've chosen not to add extra prudence for reasons Sally will discuss. She will also go through from a capital and reserving perspective, in much more depth later on inflation. I thought I'd point out that we have increased our reserves this year for excess inflation, both social and economic, having already made a specific additional provision for it in Q3 last year. We've increased our provision this year because our expectations of both the level and persistence of inflation have increased this year.
Those expectations are embedded into our capital model, into our reserves, into our pricing, and into our budgeting because inflation has an impact, of course, across the business. This issue of inflation has become a bit of a key topic this year, and rightly so. We began publicly discussing social inflation at the end of 2016, and sharing back then what we were doing about it. During COVID, we also indicated that although the impact of social inflation seems to have paused, we believed that that was temporary because nothing had happened to solve the underlying causes of social inflation, which is essentially, institutional injustice and abuse of power. We've also been emphasizing these last two years that we remain in a period of elevated risk from social inflation, and we still believe that.
We turned hawkish on economic inflation the first half of 2021 and began to adjust our assumptions again across the business accordingly. We will continue to do these things as we learn and things change. I've been saying again, and I'll reiterate it, inflation per se is not a bad thing for insurers, provided that it's properly provisioned for and thought through. The last thing from this overview is that our view of our exposures to the war remain unchanged. The increased cyber activity that we prepared for in Q1 has still not yet manifested. At the Capital Markets Day, we started explaining our capital and our platform strategy. I thought it would be useful to show our growth that way.
You can see we have good growth across all platforms, which you'd expect given the broadly positive market conditions. The domestic platforms are growing faster, with our newest in Europe growing the fastest. I do expect growth to moderate first in our wholesale platforms, and indeed, we are beginning already to take some risk off the table in some products such as D&O, where the market has become more competitive than we had originally thought when we put our 2022 plan together. We thought we would show this year two sets of rate changes. You can see, as I mentioned earlier, all our divisions do show positive rate change into 2022, but that it is moderating slightly in Specialty Risks and in MAP, but accelerating in cyber.
At a total level, prices are 75% higher than they were back in 2017. Because cyber has such an impact, I thought we would show it excluding cyber. This does two things, I think. Firstly, it makes the other divisions rate change look better, so I thought I'd show that. It also demonstrates that the rump of our business still has prices that are 50% stronger than they were in 2017, which we are pleased about, because they need to be. With that, I'll hand over to Sally. Seamlessly.
Good morning, everyone. Okay. Hi, I'm Sally Lake. I'm the Beazley Finance Director. It's really great to see you all in person. It's been a while. I'm gonna speak about my usual starter, main, and dessert, being investments, reserving, and capital, but I've added a big dollop of inflation and a slight sprinkle of IFRS 17. On IFRS 17, before everyone gets excited, we're not gonna be giving any numbers or new disclosures today. We wanted to give you some context about how we're thinking about our reserving as the adoption nears. There'll be more to come in the second half of this year and into 2023. First, let's just have a deeper look at what Adrian's already mentioned on the financials.
We've seen good growth, slightly higher on earned, and that's because we're seeing more of the effect of the strong growth we saw at the end of last year coming through the P&L. We are expecting growth to moderate slightly in the second half of the year. We're still expecting a good strong growth story, but slightly lower than what you've seen so far, in the first six months. Adrian mentioned the expense ratio is complemented by this growth being higher in the first half of the year. In addition to that, the phasing of our costs is slightly the opposite, so we are expecting costs to be slightly higher in the second half of the year. There's also a little bit of benefit on effects, but nothing too significant.
As Adrian mentioned, we're not expecting the full year to be at 33%, but slightly higher than that. Now if we go on to investments. Most investment markets are at difficult conditions during the first half of the year. With the war in Ukraine, the ongoing effects of the pandemic, rising inflation, higher interest rates, and the possibility of a recession looming have resulted in a lot of volatility. U.S. Treasury yields rose by more than 200 basis points, and global equities fell by more than 20%. Our investments returned a mark-to-market loss of $193 million, as rising yields reduced the value of our fixed income investments. However, rising yields also reduced the present value of our liabilities so that the result does not have an adverse impact on our capital position.
More excitingly, at the end of June, our fixed income portfolio yield was around 3.5%. As a result, our investment returns are going forward should be better than we have seen for some time. If we look at our portfolio, at high level, not much has changed, as you'd expect from us. However, we have made a number of adjustments which has helped reduce investment losses during the period. We've maintained fixed income duration below normal levels for much of the time, which helps limit our fixed income losses. We've reduced our equity market exposures from 3% to 1%, and we have continued to increase our exposure to inflation-protected securities within our sovereign debt portfolio. As ever, we'll continue to with our management actions as we navigate the current environment. Okay.
As Adrian mentioned, we're gonna talk a little bit more about inflation to give you, and really I'm gonna focus on how we deal with this on our balance sheet, in particular on reserves and with respect to capital. As Adrian has already said, we always think about inflation of all types that affect our book. We've always allowed for both social and economic inflation within our reserving. At the end of 2016, as Adrian mentioned, we started seeing this thing that we now call social inflation, and during 2017, we added a load for excess social inflation over and above what we already had embedded, and we've continued to do that and review it. We then started adding excess economic inflation loads at the beginning of last year.
What's really important to note is that both inflations are allowed for within both our IFRS 17 balance sheet and our Solvency II balance sheet, so they're already embedded within there. We monitor these closely. We update them quarterly. Clearly, given what's happened in 2022 so far, we updated these to reflect both the expected levels of inflation and how long we expect them to last. We did that during the first half of 2022. When you look at our balance sheet, we've already looked at the most recent view that we have, and we will continue to do so on a quarterly basis. What has that meant? I'm pleased to see, again, that we're seeing positive releases across all our divisions.
Note these are the new divisions, and we'll be going into detail about what's in there. Everyone is continuing to see positivity despite us increasing our loadings. This is after we've already increased for what we're feeling about inflation at the moment. The absolute number is quite similar to this time last year, but if you compare it to the amount of premium that we've written, it's slightly down, which is the dotted line. That's because while we're pleased to see this positive news continuing, we haven't sought to continue at the same rate because we're slightly cautious around inflation. We're looking at how we feel about inflation before we look at our reserve releases and approaching them prudently. Okay. As I said, I'm gonna mention IFRS 17, more to come.
The reason we decided to mention it is that we want to start giving context around how we're thinking about reserving. There are different rules as to how we reserve under IFRS 17 compared to the current IFRS 4. In the new regime, the risk adjustment needs to be justified in a different way to what is currently allowable under IFRS 4, where there's a wide variety of prudence levels allowed. We've always disclosed our reserve level, and we've always referenced to an internal measure that we already use, and I'll go into that in a bit of detail. Next year, under IFRS 17, this will go. Sally's favorite graph. It will be replaced by a disclosure that refers to the percentile at which we are holding the new risk adjustment.
Under IFRS 17, that level needs to reflect the risks that we are writing, which is different than under current IFRS 4. There are a number of ways which we can arrive at this, and we are yet to confirm our approach and our level. This is still to come, and we'll obviously inform you as and when we're ready to. However, our current expectation is that the level and range of risk adjustment under IFRS 17 will be lower than what is currently allowable under IFRS 4. That's the real takeaway I want to give today. Because it helps give context to what we've done on the reserve margin. To remind everyone what this graph does for one of the last times, what we're doing here is comparing the reserves in our balance sheet to our internal actuarial measure.
I'll say it again, and I'll say it slightly louder this time, that this is not a best estimate reserve. It already has an element of prudence within it. The best estimate surplus is not 5%-10%, it is greater than that, and that has never been more important. Given that we are reviewing our reserving, we've taken an opportunity to not build this reserve at this point in time because, as I mentioned on the previous slide, we are expecting that the overall range we're allowed to hold going forward will not end at the same point that this ends at the top. Therefore, it would seem odd to start building in that level. We've maintained it at a similar level, slightly lower, nothing material.
At the same time, as I pointed out earlier, our reserve releases have been slightly lower than in previous years, while we're very cautious about inflation. Finally in the trio, onto our capital position. I'm really excited about this slide. We haven't changed anything we've done here, and this will not be affected by IFRS 17. It doesn't change how we think about capital. We show the economic capital requirement needed by Lloyd's, as well as our other main capital requirement being our U.S. admitted insurance company. What's really important here is this already has a view as to what we're planning to write in 2023, and it contemplates our net premium growing by mid-teens. That's already embedded in the requirements here.
Despite doing that, because of the Solvency II profit that's being generated at the moment, we expect to be at 28% at the end of the year. We always project the capital number to the end of the year. That is slightly above our 15%-25% preferred range. I love saying this publicly, so my underwriters here, we definitely have more opportunities within the businesses, as Adrian's pointed out, and he'll go into more detail on that. We now go into detailed business planning, and it's really pleasing to see that we're in a very strong capital position as we do that, as we can seek out other opportunities to grow.
We still believe that the best use of shareholder capital at this juncture is to grow our business whilst continuing to be very mindful of market conditions at this time. With that, back to Adrian. I'll move you on.
Thank you, Sally. All right. Because we've reorganized our underwriting divisions this year, I thought it'd be useful to go through them in a little bit more depth, and to share just an overview of what's in there, and some other characteristics, and to share the P&Ls and the loss development tables. We'll obviously start with cyber because that's the one that seems to have generated the most interest. For those of you who were at the Capital Markets Day, some of these slides were in the deck there, so they may be familiar to you. To start with Cyber Risks, broadly speaking, there are two sets of products here.
Cyber insurance itself, which is our BBR, as well as a generic InfoSec policy, and then professional liability for technology and media firms, which may or may not include cyber because that's a major risk for them. You can see this is an exhibit we've shared before. The frequency by policy count and premium continues to decrease. The concern that had been echoing around the market about what the impact of the war might be has not so far manifested. Alongside that, we're increasingly confident that our cyber ecosystem is materially helping our risk selection. We said at the beginning of the year that we would look to increase exposure slightly on this book, and this is our latest forecast on that and where we think we'll be by year end.
You can see that while exposure is growing, premium is still growing at a stronger rate. This is the six-month P&L. Standalone premiums so far of just under $475 million to date, and we remain broadly on budget for this, which is about $1.3 billion for the group. We write a lot more in the second half of the year in cyber than the first. Particularly, July one is a very big date for us. Our full year 2021 premiums, for example, were just over $800 million. Good growth this year, 78% gross, 71% net. You can see we spend a reasonable amount of money here on reinsurance, about 32% of our premiums.
Part of that is on our catastrophe hedges, part of that is on our aggregate excess of loss reinsurance, and part of that is proportional reinsurance with the partners that we have that sits alongside that. Big impact this year on our loss ratios, year-over-year, demonstrating two things. First, there's been no cat, cyber cat this year, and secondly, the improvement in our pricing and our risk selection has impacted the attritional loss ratio by about 18%. This is a capital-intensive business. The combined ratio to hit our hurdles needs to be good. We've said that before. Just for guidance, our combined ratio last year and 2021 were 101% and 91%. It is a volatile class of business.
That was because of the impact of cyber criminality and ransomware that has been very well reported on. They were as low as that because the reinsurance program worked very well. Were it not for that, they would have been a lot higher. The market outlook for cyber remains strong, right? Strong demand growth should continue. We expect the market gradually to move back into some sort of equilibrium over the next 12 to 18 months. Rate changes have begun to moderate from the 100%+ we seeing at the beginning of the year, hence the year-to-date rate change being in the early 1970s. Here with the loss development tables then, a few things to take away from here. Firstly, it's not always been plain sailing in the past.
We have had spikes before when previous versions of cybercriminality have emerged, credit card theft from large retailers, for example, or systemic raiding of health insurance companies data back in 2013, 2014. Our reinsurance program has worked well for us over the years. The gross in 2013 was 67%, for example, and in 2020, it was just over 100%. We do think quite carefully about it. The table also shows, I think, that when you get it right, cyber can produce good loss ratios, but it is volatile. We used to say from a tail perspective, we'd know after about 36 months how the year would eventually fare. That was when we had a more even mix of first and third-party claims.
Third, when it was a lot of data breach activity, you'd get quite a lot of third-party claims activity from the regulatory side and from consumer class actions. By far, the bulk of our claims now are first- party, which are a lot shorter. We've got a decent idea now after 24 months, whether we think the year is a good one or not. That pattern could change again in the future as the world of cyber continues to evolve. I will underscore again that since 2013, there has been no major cyber cat activity. These loss ratios are flattered by that. On to the next division, which is our digital one.
This is where we house our automated SME business in one team and add to that the IT and operational resources that it needs to control the whole thing end- to- end, which is what we think we need in a Digital team when process and automation are so important. The main products are mostly around Specialty Risks and cyber, but we also have some contingency in there, some marine in there for some pleasure craft and so on and so forth. Our ambition is to have as much of our product set as we can for SME business in this Digital unit, ultimately. It does tend to have lower claims ratios.
Our focus for this is to invest in that end-to-end automation that it needs, which will hopefully drive down that expense ratio towards the 33% that we have as an average. Although the year-to-date growth has been lower than the group average, the outlook for this business is strong for two reasons, really. Brokers are beginning aggressively to digitize their SME business to try and drive efficiencies there, and that's really what's driving our behavior. Secondly, there is developing strong demand growth from the SME world for the sort of specialty products that we sell. This is an important investment for us. From a loss development perspective, it is relatively short tailed.
Because the customers tend to be smaller businesses, they don't tend to have the large complex claims that lengthen the tail for us, despite it being liability business. You can see there is significantly less volatility, so it's not capital intensive business. It's a relatively efficient business for us. They are quite stable, good loss ratios. You have a pretty good idea after about 36 months how the year will ultimately fare. Moving on then to MAP three pieces here, the marine bit, which writes the traditional marine products, including, you know, hull, war, energy, cargo, aviation, space, and so on and so forth. Then the second is the sort of onshore non-property first party specialty business of political accident, terrorism, life, and so on and so forth, and contingency.
Then the third being the market facilities business, so our Smart Tracker Syndicate and the new ESG Syndicate. Good premium growth for this one year on year. Lots of drivers to that. I will call out contingency as the world has begun to reengage these last nine months. Lots more events being put on, lots more shows being put on, lots of demand for our product, and a restricted market 'cause a number of carriers have pulled out of this business. So we're beginning to make back some of the quarter of a billion dollars that we lost on it in 2020. Terrorism, strong growth there too. Is that a laugh or a cry, Sarah? Market facilities also.
The 54% claims ratio, which is an increase last year, is a reflection of this is where our Ukraine losses are, right? Across the marine and the PAC bit of the book, strip those out, and you're pretty much back to where we were last year. Market outlook is a little bit more mixed. Whilst there's good, strong demand growth, some of these products are beginning to get a little bit more competitive again because the loss ratios underlying are pretty good, and we will react accordingly as we always do. You can see from the development tables that it's relatively short tail. There are some longer tail bits in here. Political risks tends to be multi-year. Contingency can be multi-year. There's some liability business in the marine business.
We've got a good idea again where we are after about 36 months. The impact of COVID is felt from 2018 to 2020 because contingency can write on a multi-year basis. Gross those loss ratios were 77%, 91%, and 88%, so our reinsurance program did partly work there. I think what this does show is the benefit of a well-diversified short tail specialty book because the loss ratios are pretty good actually over that period, despite the impact of a war and a global pandemic. I think that's a good outcome. Property Risks, two bits here. Reinsurance, of which the vast bulk is property cat, good mixture of U.S. and non-U.S., although more U.S. at the moment, because it's better priced.
Insurance, which is, runs the gamut from very small business homeowners, SMEs, all the way up to Fortune 100 business. That's mostly North American. This is obviously where the vast bulk of our nat cat risk lies, so it's relatively, capital-intensive. There is a bit of nat cat exposure in MAP, mostly cargo and energy, but the bulk of the business is here.
Good performance this year, partly reflecting quite a lot of work that we've been doing on our Property Risks business these last few years, and partly 'cause there's been fewer big losses this year for a book that, for us at least, and partly 'cause last year, the first half of the year was very much impacted by the Texas freeze, which we didn't have a comparator to this year. The 2021 expense ratio for Property Risks was quite high last year because of the loss portfolio transfer that we did on our engineering book, so it's not quite a like-for-like comparison. From an outlook perspective, this is the business we want to get more positive about.
We're encouraged by the movement in the reinsurance market, which does appear finally to be dealing properly with how to think about climate change and embed that into their underwriting. I do think this could provide an opportunity and a disciplined reinsurance market will help an insurance market as well, I think. As I've said before, if we can get the right tools to allow us to develop our forward-looking view of risk and really contemplate climate change, and the marketplace does allow us to do that, and reflect the increased complexity of property business now, there could be some opportunity here. We'll see. Our plans so far for next year do not contemplate us taking much more property risk, but we absolutely have an ambition to do so if we think the conditions are right and we are properly prepared.
The development tables do show the impacts of cat activity for the past few years, 2017 in particular, and the work since. 2018 you've got some wind losses from Japan and the U.S., and 2020 has a bit of COVID IBNR there on both the insurance book and the reinsurance book 'cause there's some outstanding claims there. As I said, the 2021 reflects not only the Texas freeze in the first half of the year but the European floods in the latter. Again, about 36 months, we have a reasonable estimation of where we think the year will end out. Moving on to Specialty Risks, which is our liability business, our onshore liability business. Quite diversified, very well diversified, and deliberately so by both geography and product. This is where our social inflation business tends to be.
It's less than 10% of the overall is highly impacted or highly exposed to social inflation. With this, we're managing very, very carefully, as well as Sally explained, loading appropriately in pricing capital and reserves. We've also triggered the recession plan for this business. Although we de-risked heavily from recession exposed business after the GFC in 2008 and haven't really put much of that business back on the books. We used to have a reasonable chunk of insurance brokers and accountants and estate agents and the like, which had a big retail exposure. We don't really have that business anymore. It's not what we do particularly well. We have less recession exposed business than we did 14 years ago.
Nonetheless, we have triggered that recession plan because of the reasonable potential of that happening. We do try to find areas of long-term demand growth to invest into, which is why you see things like healthcare and environmental and M&A in there, 'cause we like natural demand growth, especially in areas where there's some decent margin if you can do it well. There has been some good growth there, less than the group average, I think, which reflects our view of inflation, a relatively high combined ratio. We like it around 90%, if we can. Well, it doesn't need to be as low as the group average because the bulk of the investment income rests here.
The loss ratios have been elevated because of social inflation, both our estimation of it and its actual manifestation in some of the prior years. Again, the market outlook is relatively positive because of demand growth, but there are some areas that are a bit more competitive now, and so we'll be pulling back accordingly. I've mentioned D&O before, and I'll mention it again. When we look at the LDTs, loss ratios from around 56% to 70%. As I say, we like to target around 60%. You can see there, as I mentioned, the impact of social inflation, and the soft market, sort of 2015-2018. We are more hopeful 2019 and post that we've got the risk selection and the pricing better.
You can kind of see that in the initial movement of 2019 at 36 months, that we're feeling more positive about those underwriting years. Again, you can see that this is where the longest tail business resides. It takes 48-60 months for us to get a real handle on where we think the year is going. It's a long tail business for us. It's not long tail business. I mean, five years in the context of things like workers' comp, it's not long tail business. Those are the divisions. Now, onto the outlook. As I hope we have demonstrated, we are very active around thinking through from soup to nuts macro conditions, of which particularly relevant here at the moment are inflation and recession.
It's part of our MO to think about those 'cause they always have an impact on the claims environment, which is the biggest driver of losses, we think. We will continue to monitor and learn and iterate. If our views of those things change, we will continue to adjust accordingly for better or for worse. For what we know now and the expectations that we have now, we think that's fully provisioned for across our business. Very happy that capital surplus is above our target range. I think couple of things. One, it shows that we're generating capital at a healthy rate again for a business, which is great. It's a good sign that we're sort of back to where we want. It also, as Sally Lake explained, gives us more flexibility for next year. We currently plan for mid-teens growth.
If market conditions allow, we will do more than that. I will be delighted if we could. 'Cause there are lots of opportunities to grow. We've got strong demand growth across the bulk of our product set and across the platforms, particularly the domestic ones. We're hopeful about the future of things like cyber and M&A, parts of MAP, hopeful of property and so on and so forth. There's lots of things to think about and lots of opportunities, provided the conditions are right. We will, of course, continue to exercise prudent cycle management. As again, we've been saying for a while now, the risk environment generally remains elevated. There's lots of headwinds out there, not only inflation and recession, but geopolitical risk, climate change, so on and so forth. Risk is high. We're growing because we're getting rewarded for it, right?
We're able to underwrite for it, and we're able to price for it. If that changes, we will pull back accordingly. As Sally mentioned, investment yield at 3.5-ish at the moment, which is encouraging. Pleased about that. And generally feeling quite positive. I wanted to finish with a comment about the announcement from the Bank of England yesterday that our Chairman, David Roberts, will be taking on the role of Chair of the Court of the Bank of England in the autumn. We are incredibly proud for him and think that in our own small way, we may have helped him get the job. I can't help myself, sorry.
David's been a great Chair for us these last five years, which have seen a unique combination of challenges, both for the industry and for us. We would like to thank him, and I'd like to thank him personally for all the guidance and help he's given me, particularly over the last 18 months when I took over the CEO role. I think I hope he leaves us a stronger business, and we wish him every success. Our Senior Independent Director, Chris LaSala, who a number of you will meet over the next nine months, I would think, will be taking over and will be leading the search for a permanent replacement for him. With that, we'll move to Q&A.
I would like to end with the exciting news that there will be a special session in September to discuss IFRS 17, which we're all looking forward to very much. Can I ask, in the meantime, you write down the IFRS 17 questions and save it for then? 'Cause there's a limited amount we can talk about it now. All right? Great. You're gonna go this side, eh?
No, Sam.
Oh, you're going this side.
I'm left-handed.
Right.
I need to go this side.
Go for it.
Thank you. We're gonna do this standing 'cause we've decided that's a good idea.
Yep.
We're not gonna regret it.
Hi. Faizan Lakhani from HSBC. My first question is on the capital. It's a very strong position. I just want to understand how the capital requirement has developed. If you could break it out in terms of growth, market moves, and if any loads are put in or if Lloyd's have asked for any moves in that front as well. The second one is on inflation. I was trying to do sort of the back-of-the-envelope calculations on inflation to understand what the load is. If I look at the PYD compared to last year and try to adjust for that's sort of $30 million. If I adjust for the reserve margin movement, it feels like $40 million.
Would it be fair to say that the increase in loss is about $70 million for inflation, roughly speaking?
Should we do the inflation one first?
Yeah.
You do that one.
Yep.
I'll do that one.
Yep.
Yeah. All right. Looking at the movement in prior development is not a bad place to start. Right. So your numbers are... We can't disclose how much the inflation assumptions are because there's, as Sally said, there's initial expectation of inflation in there anyway, and then there's a load for excess as our things change. So what you're describing isn't the total load we've got. It's just adjustments that we have made. Looking at the various movements in prior developments is not a bad start, but it sort of assumes that nothing else has changed. You know, it's not a bad
What do the auditors or the SAOs make of that? You know, is that sort of broadly in line with the rest of the market in terms of thoughts on that front?
We haven't been SAOed since year-end, so we can't comment on that. Our auditors are fully aware of what we're doing on inflation and haven't raised anything at the half year with regard to that. It's hard for me to understand what the rest of the market are doing 'cause we're first out and I don't know what other SAO is doing. We've definitely spent a lot of time on this, so we're comfortable. The other thing to add on, just looking at prior year development, we've also got a loading on current year as well, which won't come through prior year, so that's an additional.
There's more than what you said. Again, we're not talking numbers. On the capital requirements, again, similar answer. Lots of things going on, and the requirements have gone up mid-teens. Our business plan next year is mid-teens at the moment. I would kind of broadly suggest that you know the stepping back, all else being equal, which it never is, capital requirements go in line with net premium growth expectations. At the same time, you're getting credit for Solvency II to profit coming through on the business as we earn it. As I mentioned, while it will take into account investment losses that we've seen, it will also discount unlike IFRS 4, and so you'll also get a benefit there.
They're the things that we're thinking about. Broadly speaking, you know, the requirements are growing in line with how we're growing our premium, our plans to our premium at the moment, which we'll be reviewing.
We're generating more capital.
Yeah.
Just to sort of quickly come back on the last one. Surely some of that growth is coming from rate, and that's probably not a fair reflection of exposure growth. Wouldn't your capital growth be lighter than mid-teens if it's solely driven by top line?
We don't take account of rate on day one, 'cause that's not a prudent thing to do. The benefit of rate will come through in our capital, but it won't come through on day one. That will come through over time. We are getting benefits of rate, but not on the business we're writing today. We're getting the benefits from earlier business. This is why it's quite hard to model.
Understood. Thank you very much.
Thank you.
Morning. Tryfonas Spyrou from Berenberg. Congratulations on a very strong set of numbers. Two questions on cyber. You mentioned you obviously had a very strong combined ratio. You mentioned Ukraine, Russia, claims development being sort of the trends were quite favorable because of that. Can you help us understand what is the sort of more normalized combined ratio? And then you mentioned ROE given that it's quite volatile business. What sort of ROE are you trying to solve for when you're writing cyber? And then related to that, there's been some recent news story in the press that you are amending your T&Cs to improve your systemic sort of.
Yeah
And yeah, any improvement in systemic events. If you can help us understand what does that do for capital and for growth going forward?
Yeah.
Does that allow you to write more business?
Yeah. Am I doing this? Right. Okay.
I'll do it.
There we go.
Ignore you lot.
All right. Let me try and unpick that. I think the frequency of cyber activity more generally has been subdued because of the war. Somewhat, a section of the cyber criminal world has been distracted into that war and less active elsewhere. We were comfortable with where our frequency was at the beginning of the year anyway, so we haven't needed the war to make our cyber loss ratios acceptable, but they have helped a little bit. Our expectation is they will revert it back to norm at some point because things will change, and we're perfectly comfortable with that.
The whole reason, you know, the whole reason for our ecosystem is we recognize that nothing stays stable in cyber, and we've got to adapt to recognize where the next set of threats are coming from and iterate accordingly, and that's sort of what we're doing. What sort of combined ratio do we target? It is more capital intensive than most of our business, so it needs a combined ratio that is generally lower than the average we target for the business as a whole. We sort of target about 90-ish for the whole sort of ish. So that's where cyber, we think, needs to be. It's a bit better than that this year because it's been flattered by a couple of things.
Low cats and a sort of slightly flattened attrition because of the issues around the war. You know, were all that to normalize, we'd be fine with the business currently. What are we doing on systemic? We've long talked about how we think about systemic risk, and we are comfortable with the systemic risk that we have in our cyber book at the moment. But whereas we look forward and think about a cyber market that is three, five, six times as big as it is now, and think about where those systemic exposures may grow to if they are not defined more precisely, we thought that now was a good time to get the policy forms into place that enables that business to scale, fundamentally. We've done three things with our policy forms.
One, we have redefined two exclusions, the war exclusion and the infrastructure exclusion, because both of those are quite out of date. You know, a lot's happened to what war can look like and what infrastructure is over the past 15 years, so we've kind of brought them up to date. Then we've started to describe what we think the key systemic risks are and given ourselves the opportunity to supplement those. That does a couple of things, I think. One, it allows us to manage our aggregates, but it also helps us start to define what we think is the key systemic issues are, which we think will help attract capital to the industry to allow us to hedge it at more scale.
Fundamentally, this is about being able to attract capital into the business to allow it to scale at a total level. To do that, it needs to be able to access much more of the reinsurance and alternative capital markets than it does do currently. We kind of think about property as an allegory. Right? For property to be able to function properly, it has to hedge its cat risk. To do that, it uses virtually the entire reinsurance market and the alternative capital market and ILS and all that sort of thing. I think cyber needs to be able to do the same thing. Helping to be much more precise about what it is we're trying to hedge, I think will help that transition. Need to look backwards.
Thanks. James Pearse from Jefferies. Just two questions from me. First one is just keen to get an idea of, I guess, how sustainable that high eighties combined ratio is. You know, in the near term, could we expect even better given what's happened with rate increases over the last few years? Second question is all on pricing. Looking at your rate increases in Q2, it looks like that's accelerated versus Q1. Just kinda keen to get your take on your expectation going forward, I guess, including and excluding cyber, 'cause I think you mentioned that potentially could moderate.
Yeah. I mean, if we can just go back to that showing 'cause that will give us a bit of a right. If you look at rate change excluding cyber, the Digital business has accelerated a bit this year. That's 'cause there's a reasonable chunk of cyber in there, and it's benefiting from that. Other than that, you can start to see that it's beginning to moderate a bit, and that's because the market is seeing underwriting profits start to flow in, right? You know, the market conditions in the future are gonna reflect both the recognition of those profits and other perceived risks and threats and losses that are out there or not.
You know, if we get periods of lower loss activity, we'll expect the markets to react accordingly and vice versa, 'cause fundamentally, that's what they do. Right? Assuming some sort of steady state around here at price levels that are attractive to us, we would think that if everything else remains normal, we should be able to continue to perform. But things change all the time 'cause we're in a volatile world. Fundamentally, I think the marketplace will react to things that happen, and things tend to be happening quite frequently down there, which makes it rather difficult to predict.
In cyber, it does show that the rate increases are continuing to accelerate, but that's really flattered by the first quarter. As I said, rate increases were over 100% in the first quarter. They're less than that now, and they average at sort of 70-ish. Rate increases for cyber are beginning to diminish, and that's because other insurers are beginning to see attritional loss ratios start to reduce because their cyber underwriting has increased in sophistication. Also, I think generally speaking, a lot of the corporate world has started to get better risk managed, doing better cyber management than they were 12 months ago. The risk environment is improving a little bit, plus some of the cyber criminals are distracted.
That will continue to react to that mixture of things, which makes it quite a complicated answer to a simple question, right? If everything carries on as is, sure.
For high 80s.
Yeah.
Yeah.
My answer to this is that in the past, we have achieved better than high 80s. When that has happened, we have had a significantly benign claims year in order to do that. At the beginning of a year, we've never guided to anything lower than that, I would say. It is possible, but I think you need a number of things to have lined up, and that's always been the case in the past.
Yeah. Our combined ratio, sort of 12%-16%, was lower than 90%, and that's because attritional losses were doing roughly what we expected them to, and there were very few cats. The bit between 90% and everything else is how much cat margin you get to take as profit rather than spend in losses.
Morning, guys. It's Derald Goh from RBC. A couple of questions, please. First one is just on your growth outlook. The mid-teens growth that you mentioned, what's the rating level that you've currently assumed? And also, how are you thinking about external reinsurance costs within that? And second one, could you maybe expand a little bit on the recession planning, but in specialty lines? Like, what does that entail?
Yeah. Okay. The mid-teens growth.
What sort of rate environment d oes that assume?
It assumes that it's gradually moderating a bit. It assumes that nothing much happens 'cause that's all we can assume, right? If nothing much else happens, you kind of assume that the moderation we're seeing continues essentially. That's wrong, isn't it? 'Cause something's bound to happen. Things never carry on as expected for long, but that's where we currently are, sort of. We've said mid-teens. We all said mid-teens with the hope that we can do more than that, because if things do carry on roughly as they are, we probably will want to grow more. That won't be where the plan ends up, hopefully, and we have the capital to be able to grow more if we choose to.
The recession planning essentially looks at which parts of our business are exposed to a change in claims environment during a recession, right? Part of that is where are things like theft gonna increase. The bulk of the recession exposure's in our Specialty Risks book, and it's thinking about where people are more likely to sue in a recession, and who people are more likely to sue in a recession. What we learned was, last time and the time before, is where loss activity tends to increase is where there is retail exposure. Where you're getting people and small businesses significantly impacted by economic conditions and looking to try to blame someone for it or to escape from something they committed to that they can no longer afford, right?
Which is why we saw people sue their insurance broker, people tried to stop buying houses, people sue their architect because they couldn't afford to pay for the loft conversion, all that kind of stuff, right? People not wanting to buy something that they said they would buy, all that kind of stuff, and that's the bit that we really pin down on. It's about managing our exposures to that and making sure that in advance of the recession, we've already thought about all that stuff. That's essentially what we've done. We go through each product area and think about which of those maybe have a high frequency of lawsuits in a recession. It works pretty well.
Thank you. Andrei Stadnik from Morgan Stanley. Just a couple of questions I have is, one is on property cat business. You mentioned that is something you're really looking forward to grow a bit, if I don't get it wrong. I mean, maybe I got it wrong. That's what I thought. You mentioned that property risk is something where you want to grow more than what you have done this year.
We'd like to be able to want to grow, yes.
Yeah.
Yeah.
That's the same thing. Whereas the message that we are getting from a lot of other companies is a bit different.
Yes.
People are trying to reduce property.
They are.
Because of pricing is moderating or it's not good. What is driving that? Which is the geography? Is there any particular line where you wanna grow in that?
Yeah
First would be that. Second thing is anything you have heard new on social inflation, because clearly last year there was a view that once U.S. courts open up, most likely social claims inflation will go higher.
Yep.
Any update on that would be very helpful.
Yeah.
Thank you.
Yes. Let me try and explain the view on property. Go back a few years, and property business was in danger of getting commoditized, right? Everyone was using the same models, everyone had the same view of risk, and lots of different forms of capital pouring into reinsurance. Nephila was writing a quarter share of Amwins. All this kind of stuff was happening, right? Because everyone had the same view of risk. As a specialty insurance company, it becomes more and more difficult to find what value you add on something that's getting commoditized. Roll on from that. You know, what have we found? We found actually that the models that the world is using aren't fit for purpose because they don't contemplate the fact that the world is changing.
That's why people are losing confidence in those models, and which is why capital is leaving the industry, right? Why some reinsurers are taking less risk there, because their investors are saying, "We don't trust what you're doing because your models aren't right, because the world is changing." Right? The past is no longer a guide to the future, right? That's where we sit now. We believe that as much as anyone, which is why we've been taking nat cat risk down these last few years, because we are less confident that the model's getting things right, and we don't think the risk/reward is right because the pricing doesn't reflect that. What do we need to do to be able to want to grow it more, and why would we want to grow it more?
Well, if properties become more complicated again, because property cat exposures need to be thought about differently, you can't just rely on one of the three main models to tell you what the answer is. As a specialty insurer, that kind of suits what we're good at. We should be good at being able to do things where the risk is changing or new or emerging or volatile, because you need to think about them in a different way. It kind of starts to suit us more, right? Especially if there's a price for that uncertainty, right? How would we want to try to capture that opportunity? We want to try to build models that have that forward-looking view of risk, that actually think what would.
What if, given a certain set of assumptions, what do we think climate change will look like across the East Coast of the U.S. or wherever? How do we think it would manifest it? How then do we price for that? Adding that uncertainty load we need because we have a forward-looking view of risk. That would be a very good opportunity for us to seize. It's exactly what we're doing on cyber. It's exactly what we do across a lot of our liability business when we're thinking about what we think the losses of the future will look like and trying to price for them.
What we're investing in at the moment is building those tools that take what we know from the past and think about how to apply a future-looking view of risk there that will allow us to underwrite, contemplating climate change. Because we think we'll be allowed to price for it and underwrite it because people are backing away from it, and we think it suits a more specialist world. That's the opportunity that we see. Have we baked that opportunity into our business plans yet? No, we haven't, because we haven't built the tools yet. We haven't tested them yet. We don't have that confidence to want to grow our property business yet.
We see the potential, and the reason why we talk about it is because if in 12 months, 24 months, whatever it is, months' time, we come and say we've decided to grow it's not a surprise, because we've been signaling what we've been intending to do for a while. That's the philosophy. Social inflation. Yes, you know, as the courts have reopened, we're starting to see exactly the same sort of behaviors now that we saw in 2018 and 2019 pre-pandemic. That's not a surprise at all, right? Because nothing's happened to solve the underlying causes of it, right? As we went into 2020 and 2021, we took no credit for the reduction in frequency or severity that we saw in our liability business because we thought it was temporary at best. That is what is manifesting now.
Nothing's changed. Our assumptions about social inflation, its drivers and its impacts on our losses is as it was in 2019, which was high. Hello.
Hey. I've got the mic, so I guess it's me. Kamran Hossain from JP Morgan. I have three questions. First one is on the growth plans for next year. This year, clearly cyber's been a success in terms of margins, but also growth. With your kind of very preliminary plans, you know, nothing's probably going to be right in that plan at this point.
Right.
Do you still see cyber being kind of the biggest driver of growth in that mid-teens growth plan for next year? The second question is on the combined ratio for this year. I guess, you know, 87% first half, high 80s% for the year is, you know, a great position to be in. You've had a bit of bad fortune, but good fortune at the same time. Russia-Ukraine on cyber, but also Ukraine in terms of, you know, kind of bad fortune. Would it be fair to say that balances out? I'm just trying to think about an exit number for next year, but I can work out if you just give me like better or worse and I'll work it out.
The final question is on MAP, where you sound a little bit downbeat, but a little bit realistic about the market conditions for next year. How much does that relate to kind of changing reinsurance arrangements?
Oh.
How much splits
Okay.
Within that business?
Yep.
Thanks.
All right. To go through those then. Cyber growth next year. We are very determined that we don't become a cyber insurer that does other things as well, right? It's very important to us that we maintain ourselves as a diversified, thriving specialty business. In order to do that, we need to make sure we have growth opportunities across the business. Next year's business plan does not assume that cyber takes a bigger proportion of the whole net, because you can have too much of a good thing. We've said that before. No. We think very carefully about that. You know, that's influenced by the products that we have and the platforms that we have and where we invest and so on and so forth, because you can absolutely have too much of a good thing.
Thinking about the high 80s combined ratio, there are lots of pros and cons, aren't there? Yes, we have the $50 million that we've talked about for Ukraine, plus the tail exposure on the aviation losses in Russia or potential losses in Russia. We've also had lower catastrophic activity elsewhere than the cat budgets. Generally speaking, there have been fewer claims this year than we thought that there were going to be. At the same time, we've loaded for inflation. There's lots of stuff. The overall has produced a loss ratio that's a bit better than we thought it would be, 'cause the 87% bridging to 90% is partly the expense ratio and partly the loss ratio.
The overall impact of those pros and cons for losses is a bit better than we thought. Lastly, why was I downbeat? It wasn't intended to be downbeat on MAP. It is.
You were getting tired.
No. Apart from where are we seeing the market start to get more competitive on the shorter tail business where the recent loss activity has been good. There's more of that in MAP than anywhere else. That was what was driving the sentiment. It wasn't a reinsurance issue at all. I think the reinsurance market is gonna be an interesting place in 2021, and we're fine with that, broadly speaking. As we've said before, a disciplined reinsurance market is not a bad thing.
Hi, it's Ivan Bokhmat from Barclays. Thank you. A couple of questions. The first one, just trying to understand on the reserving, and I know it borders the IFRS 17 outlook a little bit, so I apologize. I'm just trying to understand whether you've spoken in the past about PYD becoming a bit more of a component to earnings.
Now it also seems like the new accounting standard would additionally strengthen that. I mean, is that a fair assessment? If you could think of probably to quantify it in terms of, I don't know, percentage of premiums or in any way, I know, difficult question. The second one is on the capital. You said that you've taken some loadings into your Solvency II model.
Presumably you cannot see that in this 28% surplus over ECR. Can you just remind us what's the translation of that model to the Solvency II ratio and whether there's been any deviation since we last spoke about it?
Okay. This feels like me. It's a really good question, IFRS 17. The main thing to remember on IFRS 17 is that economics don't change. We get some money in and we get some money out. What we do expect to change slightly is the recognition of profit and the timing of that. You're completely right, because if we're holding overall a lower level of reserve, and we use reserve, which is wrong because it's got to start putting a pound in the jar because it will be a risk adjustment, then we'll be recognizing profit earlier. In the roundings. How that will sit between current year and prior year development hasn't been established yet. The way that the financials look are going to be very different.
That's the kind of thing that we'll be taking you through before you have to look it up, numbers and understand them, because that's the kind of thing that's gonna change. I think the thing to remember is that the economics are exactly the same. It's just the presentation and the timing of that.
At the margin.
At the margin. In the capital model, we do have a couple of solvency ratios that we look at, but we always talk about the ECR because it's the most owner-risk capital model, and the majority of our business is underneath that. We do have a slide somewhere that I'm gonna try and remember about the differences. One is that the Solvency II more standard calculation is a one-year outlook, whereas we look at ultimate for ECR, and that tends to have around a 20%-30% impact. If you had a normal solvency, you'd add 20%-30%. You then also have an uplift of 35% that Lloyd's applies to all of our capital.
The equivalent of our 15%-25% tends to be thought of at around 180%-220% in terms of Solvency II. That isn't perfect because there's lots of wondrous Solvency II things that work about that, but that's the math that I think about when I think about where we are. If you're at the top of our 15%-25% range, which we are slightly above, you're definitely into the 200s, all else being equal.
Those inflation loadings that you're taking into account.
We have taken inflation loadings within that, and we're still at above the top of that range. That's after allowing for that.
Hi. Freya from Bank of America. Just, I'm a bit confused why a change in accounting standards might change how you reserve margins internally. Is this just an accounting change and economics are exactly the same, or are you changing how much you're holding back in the real world?
My answer is yes and yes. The economics do not change. The profitability of our business is not changing. The level of margin that you hold in IFRS 4 is there is a wide breadth of what you're allowed to hold. As you entered IFRS 4, wherever you were holding was kind of acceptable, and you just weren't able to add to it. Actually the level you didn't really need to justify. That changes under 17, and there is a number of things you have to be able to demonstrate around the level that you're holding. It's around. It has to reflect the risks that you're taking.
It's around how you'd have to sell that risk in the market, et cetera. There's lots of things around it. They're the two main ones, so that when we're holding a margin, we have less scope as to where that margin needs to sit. To be clear, that happens at both ends of the spectrum. If you think that people under IFRS 4 hold a range from right down to best estimate to very high above best estimate, my expectation at the moment, again, it's an expectation, is that that range will come in on both sides. You need to hold some risk adjustment, but you can't hold too much. I think what we're trying to demonstrate is we're definitely on this side of the argument at the moment.
That's why we're trying to show that we're looking at the allowability, and we're expecting to be towards the bottom of our range at the moment.
Right. This sort of, I guess, just changes the timing of profit recognition.
Yeah.
Does that mean you'll be giving it to the stock shareholders or just keep it on the balance sheet and reinvesting for growth?
It doesn't change capital. It's only IFRS 4, so it won't be changing the way that we'll be looking at the way we look at capital.
Hi. Just a couple of quick follow-ups. As I guess on the IFRS 17, then, now we know what you meant in the footnotes to your 2021 accounts when it says there will be, I think it might use the word material, and this is significant change in profit recognition. You mean accelerate essentially.
For us, yes.
Yeah. Okay.
Yeah.
That's useful.
That's not. That's to be very clear.
No, no. It's different.
That's our position.
I understand.
Yeah, yeah.
Yeah, yeah.
Yeah.
Reserve releases will become the release of risk adjustment. There won't really be reserve releases as such.
Yeah.
Yeah. Okay.
There's a whole dictionary.
Yeah. No, I know that's different. Yeah.
Yeah.
Yeah.
Yeah.
It's all very-
I'm struggling as much. We're gonna have a before and after.
Right
All learn together.
Just a very quick one. I think the last time we spoke about capital, this plan was still to either refinance or pay down the facilities as your capital build. Is that still the plan?
So, um-
What are the thinking?
My expectation at the moment, I think we've talked about order before.
Yeah.
As long as we have the opportunity, so we have to grow. I'm happy with the financing that we have at the moment. As Adrian said, we've got a lot of work to do over the summer. We're seeing lots of opportunities, so there's no immediate plan. What I would say on the other hand is that my expectation would be that we would be looking to potentially stop posting LOC before we were going to pay anything in addition to back to shareholders. Not yet is probably my answer.
Okay. This is probably a stupid question, so apologies. I get the yield in your portfolio as of today, mark to market is 3.6%. Or is that number? I think it was the number.
Yeah.
How quickly does it earn through into the P&L? Do you see what I mean? That's your earned yield.
As we
What's the duration? Is it still two years?
As we sit here today, we're at pretty neutral duration, but we do flex that over time. The 3.6% is around neutral. It's maybe slightly higher. We're at mid-threes. Neutral is how I would think about it.
Sorry, do we have a very rapid cash yield adjustment in the second half already then?
All else being equal?
Right.
Yeah.
When did you take the duration hedge off? Is it at literally the end of June or?
When did we take it off? We've taken it off pretty recently.
Yeah. Okay. All right. The final question on Digital. Shouldn't the expense ratio be even higher?
No.
I mean, it's the growth engine. It's the area where, you know, we accept it. This is a visionary kind of, you know, division, and there's an opportunity set. Lots of people, you know. I don't know. You seem to think the expense ratio was too high, but
No.
Quite likely.
It should ultimately come down. We need to make a substantial investment to automate it. True. I think.
Steady state
Ultimately, we want it to be lower 'cause it should be more efficient. There's no point in having a digital business if the frictional costs of placing and administering it are higher than average.
I mean, we should get a. It should get a
Yeah, I'll get a company.
Right. Okay, cool. Thanks.
We're definitely investing in Digital in order to become more efficient.
Thanks.
Thanks. Dominic O'Mahony, BNP Paribas Exane. Just two questions, if that's all right. The first is, Adrian, you talked about some of the ups and downs in this half. Thinking about the guidance towards the high eighties for the full year, are there any ups and downs you're already mentally penciling in? For instance, in your 5%-10% range.
One inference to the way that you've described it is maybe that 5.9% comes down further towards 5%. Is that already baked into your expectation for H2? I guess that's helpful. Certainly for inflation, you're not expecting to add any more inflation. You're not already penciling in any more inflation load into the H2?
No. That would imply that we know more.
Yes.
The inflation loads we have now.
Yeah
Contemplate what we know and think about inflation. Your... Yes. This just assumes things carry on.
Yeah
You know, with nothing happening that's new.
Yeah. Let's be clear. It's an average attrit for the rest of the year. It doesn't assume we're doing anything further on surplus. Not to say we might not, but the guidance doesn't bake that in. Everything we know about inflation is already baked into what we've been discussing today.
Very helpful. Just second question, much broader question. I mean, you know, a couple of points increase in investment yield is a lot of money.
Okay.
I'm wondering when this is gonna start to translate into combined ratio pressure for the industry. Because if you make capital allocation decisions on an ROE basis rather than a combined ratio basis, then at some point it will feed through. Are you seeing anything in terms of how, say, your reinsurer is behaving, or capital entry into the market?
Not really. Not really. I mean, well, let's just pause for a moment to enjoy the 3.5% yield. Like just on $8 billion of assets. So I mean, we get ourselves a reasonable amount. I mean, I think, you know, ultimately there is some link between investment income and an allocation of risk appetite. Ultimately, it's a very fuzzy link that takes a while to develop. What drives prices are losses and expectations of losses, and how confident people are in their modeling. That's what drives prices. What will drive prices next year is that. What losses do people see? What are people worried about? How confident are they in their tools? Hello.
Morning. Hi. Nick Johnson from Numis. A couple of questions. Specialty lines combined ratio. What are-
Risks. Specialty Risks.
Specialty Risks, sorry.
Pounding the table.
What are the prospects for improving that to the 90% target, given what we know today about pricing and inflation? I think it's around about the mid-90s% at the moment.
Yep.
Secondly, on growth, the U.S. and European platforms, how is sort of competitive behavior evolving in those markets? Is better pricing drawing people into your markets? Has there been any change in sort of the MGA competition this year? Thanks.
Okay. Three things, I think. Specialty Risks, I think, you know, the way we've built that business, you know, with the fact that we're investing into pools of risk that are growing naturally, there should be long-term ability to grow, right? Because demand for things like environmental insurance is just only increasing, so it's easier to grow and stuff like that. We're relatively confident in the pricing environment overall, so we should be able to achieve a decent combined ratio, which is why we're growing. Having said that, this is where our social inflation risk is concentrated. There is risk there too, right? So we'll see. I think we are confident enough in that business to want to continue to grow it.
What are we seeing in the U.S. and European platforms? Overall, the way our distribution works, where we're. There's a higher proportion of retail business. We're closer to the original broker. The business is more mid-market and SME. It tends to be a more stable environment than the wholesale market here in London, which is more specialty, more large risk, and a longer chain of distribution. That's what we're sort of seeing at the moment. MGAs for us, outside of cyber and SME property, aren't a huge feature for us in terms of the competitive environment. It's not something we spend a whole lot of time worrying about generally.
Specialty line combined ratio. Cost gets to 90%.
Oh, hope
Specialty Risks.
Specialty Risks.
Two hands.
Yeah, we hope to be able to get there. We hope to be able to get there, yeah. Yes. Yes, we hope to. Okay. I think that's it. Once again, thanks so much indeed, and have a good day.
Thank you.
Thank you.