Swiss Re AG (SREN) Earnings Call Transcript & Summary

Good morning or good afternoon. Welcome to Swiss Re Annual Results 2023 Conference Call. Please note that today's conference call is being recorded. At this time, I would like to turn the conference over to Mr. Christian Mumenthaler, Group CEO. Please go ahead, sir.

Thank you very much, and hello, good morning, good afternoon, good evening, depending on where you are. I hope you are well, and thank you for joining this analyst call. So as usual, I'm going to make a few starting remarks before we go into the Q&A. Obviously, at the start, I can report that I'm very happy to have achieved all of our targets we set for ourselves in 2023, and that this allows us to actually grow the dividend by 6%. You can imagine that after the last few years, this is a very good feeling. I got feedback from Thomas and the IR team on the topics that are of particular interest and where we get a lot of questions. So I thought I'd just deepen a little bit 3 topics here. One is reserves, one is the January renewal and one on this new business [ uncertainty load ]. So first on reserving, we have obviously added significant reserves to our P&C casualty, mostly U.S. liability book throughout '23. The large majority of this was offset by significant releases in property and specialty lines. And this overall has been a theme for the last 4 years. Cumulatively, we have had near neutral reserve development at the group level since the beginning of 2020. Given the actions we took in '23, our casualty reserves are in a significantly stronger position than they were. We will publish reserve triangles on March 13 together with our annual report, and this should show prudence on ultimate loss ratios and IBNR ratios. But of course, you will be our judge, and I know that every year, several of you look at those in detail, make your assessment. So we would very much welcome that. As already shown on Investor Day, we have added reserves to most U.S. liability underwriting years, with the large majority in the years '14 to '19. We had some choices around where and how much we could add, but it was important to us to be prudent while delivering on our financial targets. On the P&C renewals, we achieved the attractive volume growth of 9%, nominal price increase of 9%. The price increases are shown as a function of claims, while volumes are based on premiums. So net, there was exposure growth in the lines we liked. You have seen, and it was much discussed, the 11% increase in loss picks. Roughly half of that reflects a prudent view on inflation, while the other half represents choices we made around updated model views and loss trend assumptions. The net negative implied price change was mostly driven by loss trend assumptions we are incorporating on our initial loss picks in casualty lines. Implied minus 2% point price change is also fully consistent with our communicated combined ratio target for '24, which is to be below 87%. Overall, the last 4 renewals that we had in 2023 have significantly improved the quality and resilience of our portfolio, but we're also making clear that we need to continue to get paid for the elevated cost of risk we see out there. A few words on the uncertainty load we introduced in the Investor Day, [ $500 ] million, which again is compatible with our net income target for the year to be above USD 3.6 billion. I mean, important to state, but obviously, uncertainty load is not reflected in the 11% loss pick increase in the January renewals. It's a separate reserving action that comes on top and that's an additional layer of prudence to the underwriting picks. So if the underwriting picks are accurate, you should see or you would expect this uncertainty load will be released over time. We're not guiding to reserve release at this stage, as you can imagine, as we aim to position the overall reserves at the higher end of the best estimate reserving range. And the reserving result will be dependent on that. So there will be yearly assessment by actuaries where we are in the range and whether we can increase. But the pure mechanical impact of the uncertainty load should reduce by about USD 100 million per year at the beginning and then going down to USD 50 million, thereafter then stabilizing just above 0. And this is, of course, a function of the duration of our claims because we're going to add it to all lines of business, some line of business, we're not very quickly. And so we will know pretty quickly whether we need this additional layer of prudence or not. Again, this is a mechanical illustration and other guidance on actual reserve releases. The priority overall is to deliver against financial targets with no separate targets for reserve movements. Maybe one word on operational expense -- operating expenses. They were up compared to 2022. A large part of that is the increase -- of this increase reflects the increased variable compensation assumptions. So we obviously missed targets last year. This year, we are on target. So this is the big part. You then have an FX part and some one-off costs related to our completed reorganizations. If you take these variables away, core costs are actually flat despite the pressures of inflation, which were very substantial. So we continue to work very hard on costs, and we'll continue to do so in the next few years. So successful '23 targets delivered, we are now focused on achieving the same for 2024. And with that, I hand over to Thomas for the Q&A section.

Thank you, Christian, and hello to all of you from my side as well. John Dacey, our CFO, is also here in the room. [Operator Instructions] With that, operator, if we could start the question.

The first question comes from Andrew Ritchie from Autonomous.

The first topic is the 11% loss cost assumption. You'll probably get a lot of questions on this. I guess I'm just trying to work out, if I take the split of your renewing book, which is now less than 40% casualty -- along the short-term lines, the loss cost assumption will be quite close to some kind of economic inflation and on casualty, clearly, it's not. But the implication is very, very high, they may be approaching 15% loss cost type assumption on casualty, is that the number? Or maybe just give us more clarity exactly what was updated on the loss cost modeling. I think you implied it was just on casualty. There was an update to the model. But it looks like there's a very high implied double-digit loss cost inflation around that or maybe building in some nonlinearity of that. The second area was we look at CorSo results. Clearly, you had strong at PYD overall. But there's a reference to updated assumptions in professional lines there. Is that beginning of something? I'm assuming that's related to post 2019 years? Just a bit of color on that would be useful.

Okay. Thanks, Andrew. I might start on the 11%. I mean, just for reference, our analysis internally shows that the social inflation, claims inflation in the U.S. has been, over the last 5 years, 16% per annum. So it's really massive. And the economic inflation over 5 years was 4%, obviously not evenly distributed. But it just gives you a sense of the magnitude, but it's not only that. I mean, this inflation all across, of course, all lines and then there's also some model changes also on the nat cat side, secondary parallels, et cetera. So it's a wide range of changes. And as I said, it's about 50-50 in terms of what we would just call inflation and model changes, and of course, some of these model change elements might also be inflationary or affected by inflation. But it's through all the lines. There's some increases in the loss picks. And yes, of course, only time will tell. As you know, we probably write all the same books as our competitors. So it's more a choice and basically our call, our initial position going into this business and then see how this will develop. But we stay cautious in view of what has happened so far. I don't know on CorSo, John, you want to?

Sure. So on CorSo, you are right the -- in the casualty lines of corporate solutions for the full year, we're reporting a reasonably high combined ratio also. This is much more related to professional liability and some of the FinPro lines, the CorSo has written. The expectation is that we've been able to reprice this book on a going-forward basis. And overall, we don't expect this to be particularly problematic on going forward with the in-force reserves now at a level which we're comfortable with. But that's the one line where I'd say we saw the need for taking some action, and that's why we've showed a full year combined ratio in casualty for CorSo up close to 110.

Andrew, I understand this might be the last time that we speak in this constellation. And while I know this is not usual in calls like that. I'd like to thank you for all these years where you have been an analyst titan, I would say, with really smart analysis, not always favorable to us, but usually right. So thank you very much for everything you have done. It was always a pleasure to work with you.

The next question comes from Kamran Hossain from JPMorgan.

A couple of questions for me. The first one is on -- I guess just thinking about 2024 and what happened this year, I guess, during 2023, losses probably came in, in some areas, a little bit lower than you had expected. You have larger reserve releases and you've kind of recycled that into I guess, the casualty reserves. If losses come in lower than you expect in 2024, you're already doing -- taking some proactive action on reserves, and we've got the [ $500 ] million number that we should expect to kind of go in. What will be the approach if things do come in, though? Will you recognize that? Will we see the upside, or will it just be kind of moved into conservatism for the near term? That's the first question. Second question, thanks for the color on the U.S. casualty side. I think one of the U.S. reporters talked about some issues in 2020 and onwards. How are you seeing kind of things developing 2020 onwards in your book?

So Kamran, let me give it a shot. First of all, it will be a nice problem to have with the losses coming below expectations. I think the important thing is that we believe that we've closed down 2023 in a well-reserved position. And you'll see this with the loss triangles that we published a month from now. But this material increase in certain lines and in certain geographies is not something that we would expect to -- has a continuous and unimpeded need. We think we've closed the year well reserved. We do have a combined ratio target next year and will -- we don't think this is an easy target to get to. We obviously have the complexity for Swiss Re to move from U.S. GAAP to IFRS and manage through that complexity. And as Christian reiterated, there's additional loading that we're going to put for uncertainty of $500 million. So I think as we go through 2024, our goal is to continue to book for the current year, a strong set of reserves that will allow us not to have any problems in the future on that year's written business, and we'll continue to evaluate whether the in-force book requires reserves or not. If it doesn't require reserve, I don't think we're going to necessarily throw lots of extra money out of it. We'll work through it during the year. On the second question on U.S. casualty for the years. Again, in the triangles, you see that we made major additions through the problematic years of '14 to '19 and in particular, with the reserving actions in 2023, probably the '16 to '19 might even be the places where the focus was. We've evaluated the subsequent years and probably did a little bit of a top of those years, 2020 continues to be an awkward year just because of the lockdowns and the question of what claims actually incurred there. But we've not left the other years alone, but -- partly because the industry pricing had already started to move and made some adjustments and partly because our book has moved, we'd already begun to reduce the exposure to large corporate risks in 2021, the amount of reserving actions was certainly much less than in the problematic years.

The next question comes from Vinit Malhotra from Mediobanca.

So one thing I noticed about the casualty renewal is there's a bit of growth you mentioned in something called structured motor transactions. And I'm just curious, is this an opportunistic thing? Because otherwise, on casualty, you're a bit cautious, especially in U.S. In the U.S., it's motor transaction, so just very curious. And just if I can add a follow-up. So my second question is a follow-up. On this 11%, just so I get it right, is there -- is it -- like how much -- or how do you read this deliberate [ prudence ] in this number? I mean what should we think about half of this is [ prudent ] or what's driving that very high number? Of course, you mentioned casualty, and you mentioned 16% per annum historic growth, inflation, social inflation. But I'm just curious if there's any more prudency there that we should also consider?

Okay. Happy to answer that. So it was motor business in EMEA, where we could grow a bit and structured is -- I mean, 2, 3 years ago, we talked about that when inflation was very -- running very hot. So structured means that typically, these are quota share. So there's quite a bit of volume. And the results, depending there's a whole corridor where the results to us is pretty similar, even if it deteriorates by 2, 3, 4 points. So it's a bit of a protected corridor around the pricing pick before you take the rest. It has to go really bad for it to be a significantly bad result. And that protects from some of -- this protected us actually the book also at the time from inflation shock. So it's not -- I mean many motor deals have that in, not every single one of them, but there was quite a bit of growth in that, and it's really EMEA, not the U.S. In the 11%, I mean, this is -- that has to be our best estimate, whether we prudent or not will only turn out -- we will only find out later. Obviously, we encourage everybody to do whatever they think is necessary, no matter how it looks. I think it is important for pricing signals, although vis-a-vis our clients and where we are. We obviously see all the other renewals reporting and as you can imagine, we probably have an overlap of 95% of the clients and probably 70%, 80% in terms of the deals we have. So I think this is always a bit difficult, this renewal reporting, as you know, because everybody uses different types of measures to it and it looks like everybody had different results. But I think it's basically the different takes on the same reality that you're seeing. And I can -- of course, we might be wrong, we might be too prudent, but it's not -- I mean, we have to make our own bet and set our own targets. So I would not suggest that in the 11% is anything we think is not needed.

And just to be clear, on that 11%, we've indicated about 50% is related to inflation, 50% is changes in some of the modeling. That modeling -- those model changes were much more significant in casualty, where we -- or the combination of model plus social inflation adjustments in casualty were a bigger number. But even on some of the property in nat cat, where we've updated secondary apparel loss models that had some impact of what we believe the required price was.

The next question comes from Bill Hardcastle from UBS.

Firstly, thanks for some of the added color here on the Q4 discrete casualty reserve strengthening. It sounds like the accident years, where the strengthening has occurred, was pretty similar to what you gave at the Capital Markets Day. I guess I've had the 22 loss pick now being about 50 points or so better than those most problematic years. And so I hear the talk about pricing, the portfolio actions that probably I have to take that into account with a 16% per annum social inflation data point you've called out. Is there any data point that can help alleviate concern here beyond the large corporate risk that can really give us comfort that we can anchor on those more recent years? Then on the second question, just trying to understand the key thought process of the Board for the dividend growth. I'm trying to understand, firstly, whether it's thought about in U.S. dollar or Swiss franc [ terms ]? Absolute or per share basis? And I guess how much consideration was given for the higher level of year-on-year growth given the confidence on the messaging it would provide?

So maybe I'll try both of those. I think the concrete information you're going to get -- level of detail that you might be asking for will show up on March 13 when we get the triangles and you get to see the movements that we've put in place across these years. I'm not sure I can, at least immediately in my mind, put the 50% increase that you've alluded to, obviously, the costing for new business in 2022 compared to the costing for business written in 2014 would be radically different. But I think terms and conditions have adjusted as well, and the level of our exposures to the large corporate, again, down 70% has made a big difference. So I guess I just have to ask you to wait until the middle of March and evaluate through yourself, whether the '22, '23 numbers make sense to -- or '22 numbers, I guess, there won't be much development in '23 in what we publish. With respect to the dividend, I think it was important to return to an increase of the dividend, and it is now a focus in U.S. dollars. The majority of our business is dollar based, and that's where our earnings come from. And we've always said that the economic earnings are material -- frankly more decisive than the U.S. GAAP numbers have been for our ability to pay those dividends. The last -- the previous 3 years, we've not decreased the dividend. It was important that we at least maintained it, even though some of the economic earnings were weak in a couple of those years, 2 of the 3. But as we go forward, I think the Board does feel confident to increase by 6% after having had this low where it was left alone. So as we go forward, and assuming that we can meet the targets of 2024, I think the Board will continue to believe that we should be increasing the dividend 6%, I'm guessing would be towards the top end of that range. But you never know. It little bit depends also where the capital base is and what we see the opportunities for us to deploy our capital into new business. But this is, I think, fairly judged by the Board to be a very positive messaging with a 6% increase.

The next question comes from Ivan Bokhmat from Barclays.

My first question would be on renewals. I just wanted to understand your view for the remainder of the year. Are you seeing some increased [ cedent ] demand that offset the capital buildup in the sector? Or could there potentially be some softening in April and onwards? And related to that, actually, for the outcome of January renewals, I mean, since we're talking about a 2% negative risk-adjusted rate change. I understand this didn't come as a surprise to your 2024 guidance, which you reiterated, but should we view that as a headwind for 2025 results. Could it be that the combined ratio bottoms in '24 and then starts to increase going forward?

Okay. I think I'm going to try this one. So the -- I mean it's always very difficult to look into the future because it's entirely going to be dependent on what happens during a year. I would judge this renewal 1/1 as a stable one. Yes, mathematically, we're minus 2 and maybe some others say plus 1, et cetera, but it's within the uncertainty, I would say, of what it is. So it was an orderly renewal at, of course, a market peak, and therefore, I think both clients and reinsurers were satisfied with the outcome of where we are. There's clearly more capacity than there was a year ago where there was panic, but that's not a healthy market if it's completely dislocated. So this was much more orderly. And I think it very much depends how the year goes, typically. So if we have a normal nat cat activity, I don't think we're going to have a lot of pressure downwards. If it's a brilliant year, nothing happens, history would tell there will be more pressure at the end of it. So it's an event dependent pathway. In terms of -- so I expect a similar trend in the rest of the year. So of course, not as easy as it was last year, where there was a panic, but orderly renewals in 1 4, 1 6, 1 7. Do we see more demand? Not necessarily. I think the clients -- some clients could, of course, place more because they wanted more last year and could not completely satisfy everything. But overall, we have not seen a huge spike in demand. I think clients are also thinking about their own budgets and how much they want to spend on this and probably some clients have absorbed some of the inflation in their own retention. And so I'm not aware of broad market increased demand.

And on the combined ratio. Could you repeat the second question, please?

Yes. So -- yes, so we had planned for -- I mean, we knew that we're going to increase our loss picks and everything quite significantly. So we have planned for something like this, what you see here as an outcome. And so this is totally compatible with our targets. So there's no reason for us to doubt that we can achieve the targets more after the renewal than before that.

And I'd reiterate the Christian's point, is the price increases that we saw in our book, 18% on January 1, a year ago, another 9% on top of that loss costs have increased, but we've got overall a reasonably well-priced book of business. And so we've not seen evidence that there's an inflection point on the combined ratio in 2024. As Christian said, a lot will depend on what the actual results in the industry will be and in particular, the level of activities around the nat cat space.

[Operator Instructions] The next question comes from Freya Kong from Bank of America.

Just want to focus on the P&C Re attritional loss ratio, which improved about 1 point in 2023. However, based on your renewals, I thought we were expecting between 1.5 to 2 points of improvement. Has this been because you've adjusted your loss cost expectations retrospectively? Or what's driven this [ slower ] under? And secondly, could you just explain the unwind of the uncertainty load again? I think you alluded to it in your opening comments. But how does that unwind? Are you adding uncertainty load just to new business in '24? Or is this all new business going forward?

So why don't I take the first, and Christian will come back on the second. So the -- in P&C Re on the attritional side, we did see nontrivial activity at sort of below the level of large loss, but still significant losses double digits, $10 million, $15 million that don't hit the nat cat budget, but we're not to be dismissed. We don't think there's any trend there. We didn't see any accumulation that was odd or any places where we thought there's a problem with any line of business. It just happened that -- especially as we came into the second half of the year, some of these added up and brought that down. So I don't think there's anything particular to be concerned. The other point is as we -- in '23, we started to lean into this additional uncertainty reserving and for the [ casualty ] lines, in particular, we're already booking additional load that would coherent with the magnitude that we would expect to put in on 2024, but that for the whole book, not just the casualty under [ treaty ] underwriting. So that's why you saw the improvement, but not maybe as big as you might have expected.

Okay. Yes. I'll try again on this uncertainty loads. And it's, of course, tricky because I can describe to you a theoretical way it works, not make any prediction about actually what's going to happen. But in a theoretical world, how this works is underwriters make other loss picks, which they have made, and then we would put on all lines of business, additional reserves to the total amount of something like $500 million this year. And so on the short-term lines, medium-term long-tail lines. As you then go into the next year, the short-tail lines, it will become more and more visible whether you need it or not. So if the initial loss picks were correct, there is -- the reserves will be redundant. And that effect, if you release them, which you would not automatically do because the actuaries would do the analysis and see where we are in the range, et cetera, et cetera. Then on a theoretical model, you would have about $100 million release, while you again do [ $500 million ]. So the net effect next year will be $100 million lower than now. And then a year later, another $100 million lower because -- and then it's depending on the shape of your portfolio. Again that sounds very mechanical. It's just for you to understand a bit the logic in real life, of course, at the end of the year. The actuaries are looking at everywhere where we are in the range. And depending on how claims wins, there might be pluses and minuses and they would be, of course, in control in what we actually do and whether we actually release reserves. But hopefully, that explains the fundamental mechanics. So it's a transition effect that becomes smaller and smaller in theory until you are at the upper end of the reserving range.

The next question comes from Roland Pfänder from ODDO BHF.

Two questions from my side, please. Could you, for the January renew provide a more economic view on what was going on. I remember last year, you had a pretax earnings number and economic view on what the results improvement would be out of the renewal. Second, I would be interested in the net cat pricing for the renewal, what you saw for your book in terms of risk-adjusted pricing?

So on the first one, on the assumption that interest rates are largely where they're going to be, I think the economic view would be this renewal is solid, but the impact would be largely flat compared to what had been a material increase a year ago. Not a bad thing, but you can't sort of expect every year to be ratcheting up on that basis. And then your second question on nat cat pricing for renewals...

I think there's always the challenge, of course, it depends, which is not the answer you want to hear. I would say the overall book is net increase in price adequacy. But there was a lot of rush to the higher layers in programs. So the more remote layers, and therefore, there was enough capacity there versus as soon as you get more into the [ risky layer, ] the more frequency layers, there was less request, and this was, of course, then leveraged by clients amongst all the reinsurers. So it really depends to the region, the layer, where it is, et cetera. But I would say, overall, nat cat is -- I mean, it's a good portfolio and further increase in quality, risk adjusted.

The next question comes from Darius Satkauskas from KBW.

So 2 questions, please. So the first question, are you still confident that your guided $500 million post-tax reserve resilience buildup in 2024 will be enough given that you've been adding a similar amount every quarter in 2023? And my second question is similar, slightly nuanced. So you've been adding to casualty for quite a while now, and you continue to add to it. So how can we get comfortable that, that guided reserve -- resilient buildup next year, 2024, it will not end up with the reserve charge given how bad loss cost trends are? So any color to help us get confidence around that would help.

Yes. So maybe -- and we don't want to be overly compulsive about this. But the famous $500 million additional uncertainty reserving that we've been talking about is for the underwriting year 2024. And this is on new business to be sure that we've got the buffers in place. A little bit to your second question, the pricing improvements that have happened in the primary industry on casualty in addition to the pricing increases we've demanded as a reinsurer. And again, that's part of 11% modeling and social inflation load that we put on the January 1 renewals, I think put us in a much more confident place for the quality and the unlikely need for any reserves for this underwriting year beyond what we're going to be booking. And whether we release the $500 million uncertainty reserve or not will be the judgment of our actuaries, some years out. Some of these lines will be pretty soon. Some of them will take 5 years to get comfortable that we do or do not need that uncertainty reserves. That's a different story than the prior year development reserving that you've seen in the 2023 and actually also some recent years on the casualty book broadly, U.S. liability and commercial motor specifically. And in those positions, we think we've ended the year well reserved. We've taken all the information that we can gather ourselves, but also from our clients, the information that finally some other people in the market are referencing in their own loss picks with their full year results here in the last month. And we think this is in good shape. Now we can't make a promise that we won't touch them, but there's no indication that I have that says there's any material need for additional reserving with what we know.

[Operator Instructions] We have a follow-up question from Bill Hardcastle from UBS.

Within the guidance, how much interest rate sensitivity can the $3.6 billion -- higher than $3.6 billion net income withstand? Perhaps maybe another way of helping us out, is at what interest rate dates with [ these struck ] out?

Look, I think we can manage a fair amount of volatility. We've got the duration on our asset portfolio of more than 5 years. We invest new funds, but there's a natural improvement that's going. You saw the trajectory of the curves and the current fixed income yield of 5% above the 3.9%, which is the yield that we noted for Q4. I think we're all fairly confident that the adoption -- complete collapse of interest rates that we can manage the $3.6 billion with the portfolio that we would have. There's -- your specific question of when we did this, we pulled together the plans, finalized them in November. So the interest rate environment that we are looking at have projections based on that. We obviously saw the dip at the end of the year, that did not overly concern us since then the long end of the curve has come back up. We also expect, frankly, that the U.S. yield curve will straighten itself out sometime in the coming 12 to 18 months. And as that happens, some of the nice gains that you've been able to get at 3 months, 6 months will disappear. But the target of $3.6 billion is not particularly dependent upon the long yields staying exactly where it is.

We have a follow-up question from Ivan Bokhmat from Barclays.

It's a very small one. On the nat cat budget, I've noted that it has come down to $1.8 billion despite volume growth in nat cats, I think as your renewal state. Could you comment on what is that related to? Is it a delayed effect of the changes to terms that we have seen last year? Or is there anything else there?

Yes. So Ivan, a year ago at this time, we suggested that with what we knew, we thought in that nat cat budget would be $1.9 billion for 2023. As we went through the renewals during the year, we reset that down to a budget of $1.7 billion, and that's -- I think we've made specific reference at the Investor Day in December 2023 was a lower exposure. But again, the actual result of $1.3 billion was below the $1.7 billion. So $1.8 billion is modestly above. Again, that's an estimate based on what we assume for the renewals in April, June and July. We'll update everyone with what the actuals are and whether we stay at $1.8 billion or whether that goes up or down by some small amounts.

The next question comes from Freya Kong from Bank of America.

Just on the attritional point you talked about, there was some volatility, some larger non-large losses fell into that ratio. So net-net, do you still expect the renewals from '23 to give you circa 3-point benefit at the end of the day, should this volatility unwind? And then secondly, just on Life and Health, was quite decently in Q4. It looks like it was acquisition costs. Was there anything special about this? Or was this driven by management actions?

So on the attritional loss, again, we saw some noise. I don't think we've used material in either direction and certainly not in trend. So -- but as I mentioned, the second component of that was we started already on the casualty book to start adding this uncertainty load. So that won't come off quickly because precisely, it was on the [ casualty ] lines, and that's not something which you'll evaluate in 12 months, but you'll rather keep for some years until it clarifies. So with respect to Life and Health, I'm not quite sure I understood the question, Thomas...

Yes. For mainly the Q4 results, the main driver of, let's say, why it was higher than previous quarters was in-force actions. That is the principal driver.

And again, these are transactions which tended to require quarters to develop together with our clients and it's not unusual for them to come in, in the fourth quarter as people try to get these done for the year-end and not have to carry them over into January or February. So that's there was a certain level of intensity of work from the client side, in some cases from our side, but we're very pleased with the resulting impacts.

The next question is a follow-up from Vinit Malhotra from Mediobanca.

Just for this -- the topic of Slide 6, [ which says U.S. ] mortality and elevated claims, just checking whether this was also seen in 4Q or was it just the 1Q? And in the earlier [indiscernible] fact that this mortality remains slightly in [indiscernible] in the population. [indiscernible] about has something to worry about? Or would it just turn down, you think, over time to normal?

Yes. Vinit, Your line was breaking up a little bit, but I think I got the core of your question, which is about the mortality in particular, U.S. mortality, which is still elevated. So there were -- so to our technical results, there was maybe 2 components. One is really the last year's Q4 '22, I mean, the year before, Q4 '22 and Q1 '23, there was a big flu pandemic in the U.S. And as we booked, we didn't have all the reports from our clients, and I think we underbooked the losses in Q4 '22. So we were -- of course, we had to do that in Q1 and Q2 of '23. So that explains part of it. But also -- I mean, also according to CDC in the U.S., we still have elevated mortality. It's much lower than the years before. So it's clearly going down. But there's still some late effects and I mean, to a certain extent, you can try to triangulate what this could be, but you're never 100% sure. So there's an element of still COVID mortality. I think I read something like nearly 3% of the deaths were due to COVID in the U.S. But as you can imagine, not everybody measures, et cetera. So it becomes more unclear. There was also a flu. There's also other viruses that are circulating. It's also conceivable or quite logic in my mind that there was secondary and tertiary effects from COVID. For example, people not going for regular screening for cancer and other diagnosis, and that could explain also some of what we're seeing now sort of a late effects, certainly the mental health effects, you have drug abuses that are higher and so on. So there's probably multiple things that are happening in the population. So it's not impossible that mortality remains a bit elevated in the next 2 years or so. I think I would expect it to continue to drop. There is sort of a momentum in it, but it would not be completely unthinkable. And certainly, we have taken some actions in regards to that. So this will be our expectation to see slightly higher mortality in the U.S. in the next years.

Gentlemen, so far there are no more questions on the phone.

Okay. With that, thank you very much for your questions. We will publish our annual report, reserve triangles and also EVM and SST results on March 13. With that -- if you have any other follow-up questions, please do not hesitate to contact anyone from IR. And with that, we wish you a nice weekend, and thanks again.

Thank you for your participation. You may now disconnect. Goodbye.