AXA SA (CS.PA) Earnings Call Transcript & Summary

Good morning, and thank you for joining AXA's 9 Month 2025 Activity Indicators Call. Our Group CFO, Alban de Mailly Nesle, will walk through the highlights from the press release that we published last night, after which we'll be ready to take your questions. With that, I turn it over to Alban.

Thank you, Anu and good morning to all of you. Thank you for joining the call today. So let me start with the key highlights of -- on 9 months '25. Overall, we delivered a solid performance with total revenues increasing by 7% to EUR 89 billion. This reflects the strength of our franchise, which, as you know, is well diversified by line of business with 60% in P&C and 40% in Life & Health and is also balanced between B2B and B2C. All our geographies are delivering a consistent execution of our organic growth strategy, which is one of the key levers of our current plan. We continue to operate at a high level of capital with a Solvency II ratio at 222%. In the quarter, the group's financial strength was further affirmed by the decision from Moody's to upgrade its ratings from Aa3 to Aa2. So let me now go through the key numbers of the release, and I start with P&C. P&C revenues were up 5% to EUR 46 billion, well balanced between the group's 3 equally sized businesses, comprising of one, large and specialty risks at AXA XL; two, small- and medium-sized risks in France, Europe and international markets; and three, personal lines. At AXA XL Insurance, premiums were up 4% to EUR 13 billion. Prices were up 0.3% on renewals with the deceleration versus 1H '25, predominantly in property pricing, which was down minus 4% in 9 months versus minus 2% in 1H. This does not surprise us given the low level of industry nat cat losses. In Financial Lines, we see the quantum of decline starting to flatten. And in casualty, pricing is plus 7%, ahead of loss trends. So pricing in the majority of our business remains attractive, and we are growing with returns in excess of our cost of capital. In addition, our retention continues to be high. So based on what we see today and how we expect the market to evolve in the near term, we believe we can maintain AXA XL's profitability in dollar terms in this plan with a number of levers. First, better investment income. We are still replacing lower-yielding assets by higher-yielding assets; two, expense management; and three, given that we are a large net buyer of reinsurance through more favorable reinsurance pricing. Our commercial lines ex-XL comprise of small- and medium-sized risks, and you know that some of our competitors classify those as retail lines. In these lines, premiums were up 4% to EUR 15 billion with resilient pricing and good volume in France and international markets. And we expect margins to expand further as higher pricing is earned through. In Personal lines, revenues were up 7% to EUR 15 billion, with good growth in both motor and non-motor. Our competitive positioning is strong. We saw close to a EUR 1 million increase in net new contracts, while we increased prices by 5%, so that will drive further margin improvement as it is earned through in the next month and year. So we have a good momentum in personal lines, both on pricing and on volumes. Finally, in Reinsurance, revenues were up 8% to EUR 2.4 billion, primarily driven by volume growth supported by alternative capital. Our reinsurance business is today well diversified with the majority of premiums from non-property lines and partnering with alternative capital will help us better navigate the cycle and manage profitability. One last point on P&C. Our nat cat experience was below our operated annual budget, in line with 1H level. And as a reminder, we manage nat cat together with PYD and discount benefit. Moving on to Life & Health. In short-term business, revenues were up 5% to EUR 13 billion, reflecting disciplined pricing. In long-term business, revenues were up 11% to EUR 29 billion, driven by strong performance in unit-linked, up 17% from successful sales initiatives across all geographies. Protection was up 11%, notably in Hong Kong, reflecting a commercial campaign as well as in Switzerland and Japan. And G/A savings were up 6%, mainly driven by growth in France and in Italy. Next, on net flows. They were EUR 5.6 billion positive year-to-date compared to plus EUR 0.9 billion last year, driven by the strong sales that I've just described and a decrease in surrenders. The improvement in net flows reflects the success of our initiatives to rejuvenate the Life & Savings business and will fuel our earnings growth over time. Moving on to new business. Life & Health PVEP was up 1% and NBV was down 1%. The NBV margin declined slightly to 4.5%. This results from strong sales in Savings and Protection with Life PVEP up 7% but that was offset by disciplined pricing and pruning measures in the multinational health and protection book, which is currently reported with AXA France business. But overall, these measures delivered a 50 bps improvement in our health NBV margin. Life saw an unfavorable impact from actuarial changes implemented in the fourth quarter of 2024 in Japan. So I want to clarify 2 things. One, these changes that we made last year in Japan do not impact profitability in the short and medium term, and we can discuss that in the Q&A, if you want. And adjusting for this change that we did last year, new business CSM on a real like-for-like basis would have been up roughly 6% at 9 months 2025. Moving to Solvency II. Our Solvency II ratio was at 222% at the end of September, up 2 points from the first half, mainly explained by 3 items: First, plus 7 points from normalized capital generation, minus 6 points of accrued foreseeable dividends and annual share buybacks. Second, plus 2 points from the sale of AXA IM, net of the full EUR 3.8 billion of anti-dilutive share buyback currently being executed. So please note that we have taken the full impact of the share buyback, though only 63% was completed as of October 28. And third, minus 1 point from the negative impact of debt redemption executed in July. So we have disclosed some additional details in our press release on the impact of the end of the transitional period in Solvency II. On a pro forma basis, adjusting for the loss of eligibility of grandfathered debt, you will see that our solvency ratio will still remain at strong levels. And from what we saw from the European Commission a couple of days ago, we are confident that the Solvency II revision, of which we will have the benefit in 2027 will lead to a significant increase in our solvency. Given the recent focus on French political risk, I would like to clarify 2 things. One, our Solvency II ratio has zero sensitivity to the OAT spread widening. Two, on the topic of French politics, we have received many questions on the various budget amendments currently being discussed in the national assembly. These discussions are similar to the ones we had last year. There is a long legislative process still ahead, and the final budget could look very different from what it may seem today. Last point on our balance sheet, given the recent credit events in the U.S. Overall, we have a high-quality investment portfolio, reflecting disciplined asset allocation over the years. The vast majority is liquid fixed income invested in high-grade ratings. We have a balanced portfolio of alternative credit with strong safeguards. It comprises largely of residential mortgages, infra and agency debt with high rating, guarantees and low leverage. We have EUR 8.8 billion of middle market lending book, which is highly diversified with an average ticket size of EUR 8 million and skewed to noncyclical industries. And we have private equity -- private equity book of EUR 18 billion, also well diversified and focused on EBITDA positive companies. So thanks to our strict investment guideline, we have extremely limited or no exposure to credits, which have been in the news recently. To conclude, so we believe we show very good numbers overall with continued growth momentum. This is driven by our diversified and balanced business model. We are executing on our plan with disciplined growth, and this gives us confidence to deliver 2025 UEPS growth at the top end of our 3-year plan target range of 6% to 8%. I'm now happy to take your questions.

[Operator Instructions] The first question comes from David Barma of Bank of America.

Firstly, on XL, can you give us an update on pricing versus loss trend by main category of business in the third quarter, please? And then linked to that, the volumes at XL were flat in Q3 compared to last year. So could you give us some color on your growth appetite in this part of the cycle and where you're seeing the best growth opportunities, please? Then secondly, on Health, can you please come back on the point you were making on the French multinational health portfolio. Can you remind us the measures you're taking there and when you expect the volumes to recover and if there will be an impact on margins once that happens?

Thank you, David. So first, on pricing versus loss trend. If we look at the various lines, -- if I start with the financial lines, which are the -- which is the line that has -- that started first to be softer. At this point in time, the pricing is down 4%. But clearly, we see a flattening of that decrease. I think we're getting to the bottom of it. But obviously, minus 4% is below loss trend. Second, in property, as I said, we see a softer market driven by nat cat. We will see benefit on the -- on nat cat reinsurance next year. And there again, so the minus 4% is below loss trend, but I would insist on the fact that property is extremely profitable. And third, in the large lines is casualty. Casualty, it's up 7%. It is above loss trend. So overall, it's true that 2 of the 3 lines have pricing, which is below loss trend. But what we believe and what I said in my introductory comments is that we have other means to offset this pricing impact. Again, reinsurance because clearly, from the discussions we have with reinsurers, prices will come down in 2026. Second, investment income; and third, expenses. But then there is also, as you point out in your second question, the volumes aspects -- and at AXA XL, we are growing in a number of places. We are growing, for instance, in property because, as I said, that's a place where the pricing is good. But we're also growing in specific areas that have a very dynamic momentum. When I think, for instance, in energy property, revenues for the first 9 months are up 61%. So you see that there are areas, I mentioned this, but we also have mid-market in the U.S., which, as you know, is one of our growth initiatives. We are up 37% on that line. So we shouldn't believe that this environment is negative because soft or lower pricing doesn't mean unprofitable. It's still a very profitable business in the vast majority of our lines. And second, there are a good number of areas where we can grow volumes profitably. On your question on health. So it's a specific business, which today is reported in AXA France numbers, which is a multinational business. And on this, there are 2 things. One, we changed some actuarial assumptions from last year, notably on mortality because we realized we had probably too optimistic mortality assumptions, and that's an impact on our NBV, not NBV, CSM but NBV because it's a short-tail business. And therefore, we are making sure that the pricing of that business is in line with our new assumptions. It's as simple as that.

The next question is from Michael Huttner of Berenberg.

I was finding it hard to find questions because you're so complete. But could you go back and would you be -- could you give us a little bit more feel for the credit mix and exposures? I know you gave us a large array of numbers. But is there -- how can I ask it? And here, I'm not even sure it's -- I mean, AXA, it looks okay. But the what could be -- if I look at it in terms of hypothetical, what could be the worse or if we imagine things getting a little bit worse, how would you -- where would you see the risk? I think that's the best. And then just very, very briefly on U.K. Motor -- no, U.K. and Ireland. We had this lovely presentation, I think, a month ago, but U.K. and Ireland, the pricing worsened in Q3. I just wondered if there's any comment there.

Thank you, Michael. So on the credit mix exposure, I believe we have a very robust balance sheet and credit quality. Just to remind you before I talk about numbers, we sold AXA IM, but we retained 300 investment professionals in AXA. On those -- from those 300 professionals, around 50 are credit analysts. So we do our credit analysis ourselves, and we give what we call the credit universe to AXA IM so that they know what to invest in when it's about lifted exposure. So on corporate bonds, starting by the liquid part, we have roughly EUR 110 billion of corporate bonds with, on average, a single A rating. More importantly, we have less than EUR 3 billion in BBB- exposure. And you know that's always the thing to look at because that's the risk of full [indiscernible]. And we are not concerned by that exposure. And more generally, we have strong guidelines, notably when it comes to trade sectors. Now your question was probably more on the illiquid or the private credit part. So on this, we have EUR 65 billion. But out of those EUR 65 billion I have mentioned, EUR 8 billion is really private credit, meaning loans to mid-market corporates with high diversification because it's EUR 8 billion, but it's on average 8 million lines. So you see that we have 1,000 lines. And again, we are giving guidelines to our GPs when it comes to the covenants that we want. We want first lien, we want certain trade factors. And they have to comply, obviously, with those guidelines. We don't give them full liberty, full freedom on the selection of the various lines. And that explains why we were hardly exposed to the names in the press over the last weeks. In addition to that, we have residential mortgages from the Netherlands, from Switzerland, in particular, for EUR 18 billion. We have CLO tranches, AAA and AA for another EUR 18 billion. And there, again, given the seniority, you would need a 40% loss on the CLO total to touch the first euro of our exposure in those CLOs. We also have infra debt for EUR 8 billion. And I'm looking at what am I missing? Yes. And we also have commercial real estate debt for EUR 7 billion. So diversified and again, very well structured and very -- with clear guidelines for our GPs. So what could be a worst case? It really depends on the scenario. But as it is today, with decent GDP growth all over and I mean, I'm not worried on our credit exposure. On U.K. and Ireland Motor, so what happened in the U.K. was that the start of the year was soft, but then it's stable. Over the last 4, 5 months, pricing has been stable in the U.K. with decent profitability and growth in volumes. In Ireland, we've seen some more pressure on pricing recently, yes.

The next question is from Andrew Baker of Goldman Sachs.

The first one, just on the French buyback and dividend tax proposals. I appreciate what you're saying it's very early, the outcome is uncertain. But are you able to just give us a sense of how you expect the process to play out? And I guess what are the key dates that we should be looking out for next? And then secondly, on the grandfathered debt, should we expect you to do any further redemptions and reissuance for Solvency II eligible debt? Or are you happy for the full impact to hit the ratio given, obviously, as you said, you've got the Solvency II review benefits coming through thereafter?

Sorry, Andrew, I'm not sure I understood your second question on the grandfathered debt.

Yes. Just should we -- obviously, you've done some work on that already for -- in the second half. Should we expect you to do more? I think you've got EUR 3 billion or so left, should we expect you to sort of do more redemption and some reissuance for Solvency II eligible debt? So how much are you willing to actually let hit the Solvency II ratio?

Okay. Thank you, Andrew. I'll start with the second because it's easier and faster than the first. On your question on grandfathered debt. So the grandfathered debt will de facto become senior on January 1. Our view is that we will be -- we'll take an economic view on that debt. Either we need it for cash purposes, and we would keep it. But at this point in time, we have plenty of cash at AXA SA level or we will repay it, but we don't have a lot of needs for senior debt -- and by the way, we don't have a lot of needs for additional subordinated debt either given the likely increase in solvency that we will have in 2027. Now on the political process in Parliament. First, what you understand is that today, the budget is voted line by line, amendment by amendment. That's the way for some political parties on both ends to put it that way, to show to the electorate that they're taking the right measures. But at the end of the day, next week, there will be a vote on the total budget law. And it's not unlikely that, that law is not voted as such. Then it goes to the Senate. And there is a back and forth between the Senate and the national assembly. And at the very end, if it's within time, the National Assembly has the last say on the budget. But if it's not within time, which is 70 days, so that it's probably until mid-December, I would suppose, then the -- we have no budget as such, which is exactly where we were last year. And the parliament votes a so-called special law, which is, in fact, a copy paste of the current budget. So that's exactly what happened in 2025. They couldn't agree to a budget at the end of '24, voted that special law, and I think it was a unanimous vote of all parties because they all realized that as opposed to what's happening in the U.S., we can't have a shutdown. That's simply not happening in France. You have a special law, which is the reconduction. I'm not sure it's English. It's a French word, sorry for that, so [indiscernible]. But the repeat of the current budget, which would have slightly, but only slightly positive impact on our budget deficit because you gain the inflation fundamentally on the tax thresholds that you have. So that's the process in the next 2 months. And so we'll see where we land.

The next question is from Thomas Bateman of Mediobanca.

The first question is just on AI. I think it's a year since you presented on the topic. I was just wondering if you could share roughly how much you're investing and how you evaluate capital allocation to these type of projects versus alternative uses of capital. And then the second question is just on leverage. I was just wondering what's the best way to look at it? You obviously quite comfortable under your own gearing metric, but we seem to be quite close to the limits on Solvency II. I was just wondering what you think is the right lens to look at leverage fee.

Thank you, Thomas, for your 2 questions. So on AI, obviously, AI is very much front and center of our IT investments. And to give you an idea, approximately 40% of our IT spend is investments. That being said, to be ready for AI, you need to have prepared your systems from the ground up. You need to have moved to the cloud. You need to have the right data set. And so that's also where we are investing in. We are investing in getting out of mainframe. We have almost finalized our transition to the cloud, but you need to be ready. So this is -- there's a lot of traction, obviously, and we're getting ready for that. But I wouldn't say that it's only AI. It's everything that you need to get to the use of AI. And when we say AI, it's predictive AI, generative AI and agentic AI. On leverage, you're right that with the debt that we have today, the one which is grandfathered and the one that we've just issued, we have saturated our Tier 2 capacity to the point that, as you may have seen in the press release, roughly EUR 300 million of Tier 3 is now a haircut. But with the end of the grandfathering, the current EUR 4.2 billion, of which you saw that we repaid EUR 1.2 billion in October will disappear from the Tier 2 and Tier 1 pockets, and that will give us additional capacity if we want to issue in the next years. But as I said previously, given what we have issued, given the uplift in solvency that we have in '27, we see little needs, everything else being equal, to issue further debt.

The next question is from Farooq Hanif of JPMorgan.

I hope you can hear me with my new headset. But I wanted to go back to the point you made about investment income and expenses as a way of maintaining profitability at AXA XL. So of these 2 things that you mentioned, I mean, what's new? I'm guessing when you refer to investment income, are you talking about the fact that yields remained higher than you planned. So you're not talking about taking more risk. And then on expenses, can you talk a little bit about what would go beyond your current plan so we can understand what to add going forward? And then my second question is just on your credit exposure. I believe that a large proportion of this is in participating funds. So could you just talk about the shareholder versus policyholder split roughly so that we can have some extra comfort on that?

So on XL investment income, the short answer is, no, we're not taking more risk, but you have 2 simple factors. One is we're simply growing the balance sheet with revenues and growth in balance sheet simply means higher investment income. And second, it's the fact that we do have higher-yielding assets, plus I believe, but I may be wrong on this, that we probably have seen the trough in terms of PE distribution. You start to see a rise in M&A in the U.S., and that could lead to better distribution from PE funds going forward. I'm not counting on this. The other 2 factors are sufficient, but it could come on top. On expenses, I think at some point, there is the projects that you have and the plan that you have. And by the way, we will keep investing in our initiative in the U.S. where we want to accelerate in E&S and mid-market. But when the market becomes softer, there are also a number of projects that you can postpone, you can reduce and so on. So you have flexibility. And I believe there's also further transformation in AXA XL on the expense side that we can think of. It will take a bit of time. Maybe it will not be all done in '26. But for the next period, I believe we have some room to reduce expenses at XL and notably through AI. On credit exposure, so we disclosed in our financial supplement the part which is in participating fund. On average, it's 60%. When -- and I think you have some details of that in the financial supplement.

The next question is from Andrew Crean of Autonomous.

Two questions. Firstly, on AXA XL's EUR 13 billion of revenue. Could you give us a bit more of a split in monetary terms between the different areas? Firstly, how much is U.S. and how much is non-U.S. of that? And then secondly, when you go into the U.S. side, could you give us a sense as to how much of that is financial lines, property, large corporate and then the bits which you're interested in sort of the mid-market and the specialty lines? And then the second question is, you've talked about the benefit of Solvency II reform in 2027. Could you give us some sense as to what you think the points benefit of that is and compare and contrast that with the potential impact on your solvency if you did a burn down or recreated the global financial crisis of 2008, '09, what would that do to your solvency?

Okay. So on the XL, as you know, we don't disclose a lot of details, but I would say the following: One, it's roughly 50% U.S., 50% non-U.S. in terms of revenues. On volumes, I think the important thing to have in mind is the flexibility that we have. I'll give you an example. You saw that financial lines had a soft market over the last 3 years. Financial lines, probably 7 years ago was EUR 1.3 billion revenues. It went up to EUR 2.5 billion with the hard market. it's down to EUR 1.2 billion. And you didn't see it overall because that was compensated by growth in other areas. So that's the way we want to manage AXA XL, growing in some lines, decreasing in some others where it makes sense. And I would insist again that we want to grow because most of our lines are still in a very profitable territory. On the Solvency II reform, so what we -- we are waiting for the final confirmation of the trialogue, as it's called, to have the final number. What we've said for a number of periods now is that the impact should be between 10 and 20 points. At this stage, it's probably at the upper end of that range. But I want to be cautious on this because it's not finalized yet. Things could change, and we need to take that into account. And so that gives you a view on how it would compare, for instance, to the points we would lose on the financial crisis. As you know, it's 30 points. To be very transparent, I have not yet recalculated what the sensitivities would be in the new framework post Solvency II revision.

Okay. Alban, you can't do any more on the U.S. business in terms of looking at the different areas other than what you said about financial lines. Is that you're sticking?

I mean at this stage, again, we need to be fair in our disclosure, and we are not reporting more detail than this publicly. So I'm sorry for that.

The next question is from Dominic O'Mahony of BNP Paribas Exane.

Thank you for such detailed responses to the questions thus far. I've got a few left. Just coming back to the XL investment opportunity. I suppose I had assumed there will be some headwind from discounting from here given some of the movements, especially in the U.S. Is that fully offset, you think, by the net financial result dynamics. That's all sort of within your thinking on the sort of the flat dollars of earnings. Second question is just on the retail side, and I suppose retail and SME side, there's been some commentary that frequency has been quite benign across some European markets. I was curious to hear whether that's been your experience? And if so, what do you think might be driving that? And then thirdly, just coming back to the Solvency II reforms. The percentage points could, of course, be quite large. Do you think it changes anything in the real world for you? Do you think it actually impacts your remittance capacity when you think through what sorts of entities are being -- are going to see the benefit and what drives the remittance capacity? Do you see any impact there? And/or do you think it changes your incentives on asset allocation? Do you anticipate any mix shift in that?

Thank you, Dominic. On the XL investment and discount. So as you know, discount has 2 components. One is the rates. And so you need to look at, on average, the 5-year rate. And there is some volatility in that rate coming from the tension between budget deficits on one hand and lower growth on the other hand. But our belief is that it will remain at a high level compared at least to history. So that's the first point. But the second point is also the duration of your current year claims. And you saw that in half year, for instance, we kept a strong level of discount benefit because the mix of lines was skewed to more long-tail lines, which is normal because when you see that casualty in the AXA XL has price increases of 7%. By definition, you grow your casualty lines faster than your short-tail lines. So that also explains why we believe that the discount benefit should remain at a reasonable level and therefore, should not be a strong headwind going forward. On frequency for retail, yes. I think it's true that we have better frequency in Motor than last year, simply also because it was less rainy in Europe than last year. That's as basic as that. And so the -- we do see an improvement. And that's why when we see this plus the 5% price increase on average in Motor, we believe that we have some room to improve margin. On the Solvency II reform, what I would suggest is that when we disclose our numbers in February, we give you the full view on how much it will really -- what it will really mean both in terms of solvency and in terms of remittance. I think at this stage, it's too early given that we don't know yet what the outcome will be. And sorry for that.

[Operator Instructions] The next question is from Michael Huttner of Berenberg.

Just to reinsurance buying, I'm working out that it's a EUR 300 million benefit. I just wondered if my math is right. So I'm assuming very rough numbers, EUR 2 billion kind of protection, so EUR 3 billion premiums, and my numbers are probably incredibly wrong, price decline of 10%, so EUR 300 million. And then on Japan, you said you'd -- if we asked, you'd explain why the lower margin is kind of not particularly meaningful for earnings now.

Sorry, Michael, could you repeat your first question? I'm not sure I got it.

It's just on the benefit of lower reinsurance costs. I tried to work out a number. I mean, it's a bit -- I'm assuming EUR 3 billion total premiums, 10% decline at EUR 300 million, but I have no idea. I'm just making it up. I just want to...

So the -- I would say, what we see is clearly a market where prices are coming down, notably on nat cat. It's also positive for reinsurance buyers on long-tail lines, but the more important impact will be on short-tail property. It's probably, it's a bit difficult to say at this stage how much it will be, and it's still being negotiated with our reinsurers. I would rather not paint that discussion with reinsurers, and I will tell you in February what the outcome is. But net-net, there is obviously a strong benefit for us. On Japan, so the actual changes we made in Japan were the following. We realized 2 things. One was that the -- that surrenders were slightly higher than what we had projected. So we increased our surrender rate in our projection. As such, it has no bearing or hardly any on next year's earnings because what it means fundamentally is that you will have fewer policies in 20 or 30 years, that's an impact, but not that you will have lower margin next year. And second, there are some products for which we -- that we modeled better, notably introducing options and guarantees in the modeling, so TVOGs. There, again, when you unwind your CSM, you unwind those TVOGs and that has no bearing on your earnings. So we -- the new business CSM last year was impacted because of that in a risk-neutral environment, but with no impact on our real-world earnings short to medium term. And compared to those numbers, we grew new business CSM this year by 6% if you adjust. So we believe it's a good performance.

The next question is from Andrew Baker of Goldman Sachs.

Just a quick one on AXA XL and the casualty pricing that you're talking about. Are you able to give us a little bit more color on what specific lines you're writing here? I'm just -- I guess I'm just trying to square the 7% price increases being above loss cost trends. So any additional color there would be helpful.

So we are obviously writing a lot of lines, both in U.S. and international. We do primary and excess in the U.S. We do construction casualty and so on. So there is a different dynamics in those lines. I would say it's probably lower price increases in International Casualty and higher price increases in excess casualty in the U.S. That's -- and overall, it comes to 7%. But the underlying dynamic of the loss trend is obviously different if you compare Europe and U.S. and even within U.S. primary versus excess.

Okay. Ms. Venkataraman, there are no more questions registered at this time. May I turn it back to you for any closing remarks.

Thank you. Thanks, everyone, for joining our call. If you have any follow-up questions, then please don't hesitate to reach out to Investor Relations. Have a nice day.

Thank you very much.