Okay. Good morning, everybody. Welcome to Morgan Stanley's European Financials Conference. For those of you who don't know me, my name is Hadley Cohen. I run the insurance research team here at Morgan Stanley. I'm delighted to be joined by Guillaume Borie today from AXA. Guillaume is the global head of strategy, underwriting, risk, finance, and technology. Guillaume, welcome. Thank you very much for joining us.
Thank you very much, Hadley, and good morning, everyone.
First of all, can I say congratulations on the new role.
Thank you.
Only a few months in. Maybe to start, if you can sort of say how things are going so far, what you've been particularly pleased with, what you've been particularly surprised by. Yeah.
Well, look, I have to say that I've been lucky enough to join in this new role at a time where the group, AXA, is in very good shape. You saw our earnings and results for 2025, delivering really at the top end of all of our targets. Strong top-line growth, 6%. Strong UEPS growth, 8%. Also a clear confidence message that we gave you when we released our results. We are committed to deliver at the top end of our range for 2026 again. The group is in very good shape, and I tend to say that all businesses today are really firing up, and we saw very positive dynamics across the board.
When I look at that, if I have to think about not surprises, but things that over the past few weeks I've carefully looked at, it's the very fact that starting from this strong position, starting from very well-performing businesses, I do believe that there is still some upside for all of us. That across the board, if we are extremely focused on a couple of top priorities, we have the ability to keep growing earnings at a sustainable pace year- after- year. When I look at those priorities, I would quickly mention four this morning, where I would like all of us across the group behind Thomas to really focus ourselves. The very first one is organic growth. We can do even better on organic growth, growing volumes. Why so?
Because we have very strong margins across the board. We have really over the past 10 years, managed to develop a pure insurance player with very strong level of technical discipline. It's a good platform out of which we can grow. The second top priority for all of us should be to accelerate the scaling of our artificial intelligence solutions. Accelerate really a high-quality AI agenda. I see that as a strong opportunity, and I guess we will have questions on that later. A strong opportunity to improve all the elements of the P&L. We can come back to it. The third element is that I still see some room for improvement on the efficiency side. We can improve efficiency discipline across the board.
Obviously, within our operating companies, more efforts on cost efficiency, more effort on automation onshoring, but also at the corporate center level on our holding costs. Last but not least, we will continue to maintain extremely strong capital discipline. Obviously around our distribution policy, 75% distribution, but also around the way we deploy capital locally in order to fuel the organic growth and in order to fuel investment in artificial intelligence. You see all in all, I think the group is in very good shape. All businesses are performing well, and there is room for further improvement around those four priorities. I'm very positive.
Excellent. I think we'll come onto all four of those priorities in due course. Before I do, we're all for audience participation here at the conference. We do have a polling question for you, which should come up on the screen in due course. That is, what is currently the most important investment debate for AXA shares? Your choices are commercial pricing outlook and XL profitability, cash remittances versus shareholder payouts, capital management, so M&A versus share buybacks, lower EPS growth targets versus peers, French political risks, or AI disruption risks and opportunities. You have 10 seconds, which should start now. Interesting. Commercial pricing outlook, XL. It's more evenly spread than I was expecting.
I can see Anu was taking bets in the front row as well. Maybe if we start on P&C more broadly, given that's where the majority ended up. Before we get onto XL on European retail side of things.
Yeah.
Pricing has been very, very strong.
Yeah.
It's been a useful tailwind for you, for the last couple of years. How much more momentum do you think there is in those lines of business?
I think on the retail book in continental Europe and to a certain extent in this country, in the UK, there is still a positive pricing momentum clearly. Obviously if you look at the absolute figure, it's going to reduce and it has already reduced in 2025, and it's likely to be the same trend in 2026. The absolute level is not that interesting. The question is the relative level against our claims trends and against also CPI. The reality of the retail market is that we have been permanently able to really sustain pricing above CPI for the very fact that first, this is more or less the claims trends in this business.
Second, let's not also forget that we benefit across the board from relatively stable if not reduced frequency of the risk that we insure. Again, the pricing environment in my view in retail is positive. Therefore for us, I see that as a good opportunity first to further expand margin. I do believe that in retail SME, in P&C across the board at AXA, we still have room to further improve technical margin, and that you will see that in 2026 and in the coming years. On top of that, it's again, a great opportunity for us to seize this opportunity, this pricing environment, this conducive environment in order to grow our book. For that we have very strong assets, high quality distribution networks that we further expanded in 2025, good partnerships in several European countries.
Look at what we did in Spain with Correos. Look at what we did in the UK with the Lloyds Bank. We have solid distribution partners, proprietary, non-proprietary. We have very strong direct franchise, the leading direct franchise of all the international players in insurance, and we are permanently investing in customer retention. I think that in the current environment when you are best in class in terms of technical margin and pricing quality, the opportunity is also to further improve customer retention. Are we going to reduce customer churn? That's for me one of the challenge we have, and I believe an opportunity in the end to grow our volumes.
Very interesting. Linked to that, how can you talk a bit more about this customer retention?
Yeah.
What you're doing to sort of focus on that, be it with regards to cross-selling products, or what have you. Presumably this is all a key part of the organic growth story by reducing the churn.
Yeah, absolutely. That's not only a key part of the organic growth story, but that's also in my opinion the best position to be in. Because by definition, the more we will retain our existing customers, the more benefit you will see both on the top line but also on the bottom line. Because when we retain customer, it's better margins. We know that. We live in an industry where attracting new customer is extremely expensive. The more we can do to best retain our customer, the better it will be for all of us. Where do we focus ourselves? I see really three areas. First, it's a matter of pricing adequacy. Are we able in our retail books to really price each and every customer at the right level? It's not an anecdote.
Over the years, many of the players in this industry have the habit of pricing across the board and pushing pricing increases at every renewal that are exactly the same for all their customers. We see that a lot in our local markets with the small players in the markets, for example, Mutuals. We have a much more personalized approach to pricing, including through AI. With AI, we now have the opportunity really to revisit the pricing at the level of a single policy holder. You need to have a better offer at renewal for the best customers in your portfolio. That's the first element of customer retention. To a certain extent, in my view, it is the most important one. The second top priority is around distribution efficiency. How do we make sure indeed that we invest in cross-selling?
Because we know that obviously the highest number of policies per single policy holder, the best retention you're gonna get. But also on distribution efficiency, making sure that they know how to defend their existing portfolio. It's the quality of the solutions we give them, including in terms of CRM solutions. It's also the quality of the products and solution we are offering on the platform. Last but not least, on customer retention, obviously quality of the service we provide. We have invested a lot on those questions. You see that with the high quality of our NPS score across all of our retail operations. We need to continue to invest on that.
There also I see strong opportunity with AI. In order to be much more efficient in the way we pay claims, the fastest possible, and in order also to deliver personal advice to our customer, in particular through their smartphone or all the platform they are now using.
We'll come on to AI shortly. You know, I think the organic growth story and the strategy there is quite clear on the retail side. If we move maybe to the commercial-
Yeah.
side of things, you know, there is clear pricing pressure in that market right now. How are you thinking about top line versus bottom line for the XL business and commercial lines more broadly?
First, I think it's important to keep in mind that when I look at the P&C business of AXA, it's EUR 58 billion book. Out of this EUR 58 billion, 2/3 is retail SME mid-market, mostly in Europe. Those 2/3, they're gonna follow the trends we have just discussed before. Good pricing environment, still opportunity for us to further expand margins. It's true also in SMEs, and good opportunity for volume growth. There again, it's not the same trend as the trend we are seeing on the large commercial book of XL, which is 1/3 of our portfolio. On the XL portfolio, yes, the environment is more competitive. It's a fact. Can we say that there is one cycle across the board? I don't believe so. I think that much more than focusing ourselves on pricing cycle, we should discuss pricing adequacy.
We do have opportunities to grow our portfolio at XL under very good margins and with good profitability. We have profitable growth opportunities. XL is not just a book in the U.S., for example. XL is 26 countries, 400 products, well-diversified book of business. When you look at the split of the business, you have casualty, you have property, you have professional lines, you have specialty. The trends are very different from one book to another. The name of the game for us is diversification and agility. Over the past 3 years, we have prepared ourselves for this environment. How? By changing the way we incentivize our underwriters across the world, by developing a much more agile portfolio management process within XL under the leadership of our CEO, Scott Gunter.
Today they are able to grow where they want to grow because they see attractive margins. They are able to reduce the exposure where they no longer want to have this exposure because we are not able to get the right capital returns. How do we look at the development? For each and every book, and again, 26 countries, 400 products. There is not a single policy or a single view of the pricing environment. For each and every book of business, are we able to grow above our capital hurdles? When the answer is yes, we need to grow, and we will grow. When the answer is no, we'll reduce our sensitivity to the business. That's why also when you look at the book of business 3 years ago, it was not the same as today.
The mix will keep evolving in the next three years. Can I tell you today exactly where it will be in the next three years? No, because again, it's a matter of agility in the underwriting teams. What I can tell you though is, one, we will protect our margin. So we grow when we see profitability. Second, that's what you saw in 2025. Even if overall, when you look at the XL Insurance books, you see stable pricing. It was 0% price effect for XL Insurance in 2025. You have seen growth in volumes 3% on average. So that's exactly the indication of what I mentioned earlier. Third, it's not just stable technical margins. We are also able to work on the other elements of the P&L. When I look at XL Insurance, what do I see? I see permanent focus on expense discipline.
You had an improvement on the expense ratio in 2025. We will keep working on it in 2026 in order to have high-quality earnings. Second, we are benefiting from the good reinsurance environment. We had very good reinsurance pricing at renewal this year, in particular on the property side. Obviously, you will see this impact in the P&L. Last but not least, while not changing our asset allocation strategy, we are improving the investment income. Why? Because obviously we are investing at a better yield, as we speak, given the interest rate environment. All of this is conducive, in my view, to earnings growth at XL. Therefore, when I look again overall at the situation, I see the commercial book of AXA XL as a good opportunity for us to further expand EPS growth for the group.
Very comprehensive. Thank you. If we maybe change tack slightly and go on to the hot topic du jour, the AI-
Yeah
Conversation. One of the comments that you made on the results call, which I was particularly interested by, was that you think you've already seen a 1 point benefit on the claims ratio from AI implementation. I was just wondering how you measure something like that, and also how you think about the scale of opportunity still to come from that respect.
Well, first, I would say that I really see AI indeed as a great opportunity for us. When you look at the insurance industry, the reality is that over the decades, we have been struggling, all insurers, with one basic element of our model, which is that on the one hand, you want to obviously industrialize as much as possible your processes to get more efficiency, to be more agile. On the other hand, if there is one business where ultimately the end customer is expecting a very personalized answer, it's insurance. I do have the view that it's much more the case with insurance than with any kind of other businesses starting, for example, with banking. Much more because why you want to really personalize solutions in terms of underwriting. It's true obviously for companies, but it's true for retail consumers as well.
Also at the point of claims, that's when you expect your insurer to be extremely personalized in the way the insurer will answer to your claim. It is a clear tension in our model. With artificial intelligence, we finally have the technology to solve both issues at the same time, really industrialize at scale and really delivering personalized answers along the value chain. That's why I see that as a strong opportunity. It's a technology change. As with any kind of technology change, the name of the game will be, are you able to invest enough? Do you have the scale to really put this innovation available to anyone? Do you have the right distribution networks? Do you have the best teams to make it happen? Those questions are not going to change.
It's a basic, I would say a basic metaphor that I will use now, but you take the best motor engine in the world. If you put it in the hands of the worst car manufacturer, it will yield absolutely nothing. We have the same question with AI in the insurance industry. It's a wonderful technology, but it doesn't mean that you can succeed whatever your size, the quality of the people or the quality of your competitive moats. I do believe that with AXA, we have the right assets in order to make it happen. We have invested a lot over the years in the quality of the technology platform, fixing our legacy issues.
We have a gigantic amount of data that is available in our systems, and we have very strong technical people that are able to train the AI in order to get the best answer. You need to have this intellectual knowledge in your company if you want to make it happen. That's the opportunity. I will answer your question on where I see the impact. Very quickly. Impact is on four dimensions in my view. First, on distribution. You can improve the productivity of your distribution with AI. We are already doing it in several countries. One example in France, now 2,000 tied agents are helped on a daily basis with an AI assistant to get the best answer possible to the customer they have in front of them. It's improving their productivity at least by 10%.
We are still working on the measurements, but we really see improvement in the productivity. That's gonna be more top line. That's gonna be also better answers to the customer needs. I make the assumption that it will improve the loss ratio ultimately and the quality of the technical excellence. Second area, pricing. That's where at this stage we have already seen the biggest impact. That's the 1-point improvement in the loss ratio that we mentioned in our call when we released earnings. Also, we are investing again in the quality of the pricing, dynamic pricing, making sure that we use much more data points. I will give you a basic example. We recently acquired Prima in Italy. To price any kind of risk, they are using 130 data points. The second best performing business at AXA is our direct business in France, Direct Assurance.
They are using 80 data points. You already see a gap. The more data points you're gonna get, the best pricing obviously you're gonna have. When I look at deploying that kind of AI tools, now we are working on motor pricing. We have a tool that we are deploying across market. It's live in France on the book of AXA France, not just the direct business. It's live in Spain. The nickname internally is Photon. We deploy this tool now in other geographies. When it's deployed, we see this 1 point improvement in the loss ratio against the books where we have not deployed it yet. This is the 1 point that we mentioned. On pricing, it's already working. The more we will deploy it, the better you will see the impact on the loss ratio. Third area of focus on claims management.
There also, we need to work on the two legs. Speed. With AI, we will be much faster in the way we settle claims. We all know that in our industry, the faster you are on claim settlements, the lower the cost of the claim. So that will be a benefit on technical excellence. On top of that, with AI, we are already seeing improvements in fraud, abuse, waste detection. It's already yielding results across the board, in particular, again, on the retail side. We will accelerate our effort. That will help us improve further the loss ratio there also. Last but not least, fourth area. In general, yes, we expect an impact on productivity coming from AI. Productivity in terms of speed.
Our people are gonna be able to work swiftly through AI tools and productivity also in terms of time to market, in particular on the technology side. Let's not forget that a bulk of our costs are coming from technology costs. If I'm able to use those costs much faster, I'm gonna put solutions on the market that are more innovative, more adapted to the customer needs. It will drive accelerated top-line growth while lowering my expense ratio. When I look at AI, again, at AXA, we believe that it's an impact that will go far beyond just the expense ratio. That's important because the expense ratio, it's 10% of the P&L. If I can impact the rest of the P&L also with AI, the value we will unlock is gonna be far greater for all of us. That's good news.
Again, and I will conclude there, the name of the game for all of us will be to make sure that we are fast enough in deploying this technology at scale. There, I do believe that we have great assets in our hands, starting with the quality of our people, the quality of our brand, and our local scale across markets with the distribution networks. That's what will make the difference on the ground.
That was an incredibly comprehensive answer.
Sorry if it may be a bit too long.
No, no.
I'm passionate about that one.
You've already answered a lot of my follow-up questions. You know, I think there is and you've touched on this around your scale, your speed to market and what have you, but there is a lot of debate in the market right now around the influence of LLMs from a distribution perspective, and what have you. It sounds like you're well on top of that. Do you perceive any real threats from AI, or do you think you're so far ahead of the competition that actually it's not really a concern for you?
Well, of course, you need to be permanently obsessed with potential threats, and you cannot afford to rest and think that you're on the safe side. We are extremely mindful of those evolutions. Again, I see that much more on the opportunity side than anything else. Again, look at the distribution question. It's a change in the way we will interact with our customers. Insurance is not, has never been just a front end. It's a lot of capabilities in the end. The question for us is how we use AI in order to transform all of our capabilities. If customers going forward want to interact with us through an LLM platform, the question for AXA is, how do I make sure that I'm the first one on this platform?
That's the question we had to solve 20 years ago when price comparison website started taking off, and we have solved it. In many of our markets, we generate 25%-30% of our retail business through that kind of platforms. Look at continental Europe. We have the best tied agent networks across Europe. In my opinion, it's really the best networks. They generate today a significant bulk of their business through Google price comparison website. They capture the customer there with the help of our technology stack, and then they transform this lead into a multi-equipped customer where we create a lot of value. LLMs are not going to change these fundamental points. The question is where you access the customer. If LLMs are taking off, we will be there, and we are already investing on that.
We have deployed an engine across all of our operating companies in order to help our marketing and distribution teams assess where we are positioned in the LLMs. When we started deploying that, guess what? In many markets, we were down the road really like number 5, 6, 7 when you were asking on ChatGPT, for example, where can I buy my motor insurance? We are catching up very quickly. I want us to be permanently in the first rows, and that's exactly. Again, today, how many of our customers are coming through those LLM platform? Less than 1% of our new business. It's a minimal question today. Tomorrow, we will adjust. That's the way we look at it. We need to be agile enough in order to find the customers where they are.
Again, it will not change the many assets you need to have if you want to be a great insurer.
Thank you very much. If I think about bigger picture, you know, we're coming towards the end of the current multi-year plan. You're expecting to deliver at the upper end of the 6%-8% annualized EPS growth. I mean, how should we think about that 8% effectively EPS growth going forward? The reason I ask that is a couple of things. One, you're painting a very positive picture around how much growth there is still, how much efficiency improvement, and how much momentum there is in the business right now. Equally, I look at your current 6%-8% target, and it is towards the lower end of what your large cap peers are targeting as well.
How should we think about the momentum of the business given you're delivering 8% growth right now going into?
I don't have any scoop to share with you today. We will all meet on September 21st for the next plan. When you look at us, I believe that the way you should think about us is a question of predictability and consistency. That's what we are committed with Thomas and the management team. That's what we are committed to. Predictability, consistency, no surprise, being sure that we deliver year- after- year. I do believe that this is what we have delivered over the past three to four years extremely consistently. Look at the quality again of the earnings in 2025. We gave a clear guidance on 2026. For the next plan, let's meet on September 21st.
I was hoping for a bit more than that. I mean, if we stick with the sort of bigger picture and from a balance sheet perspective, as well, you know, your solvency right now is very, very strong, and it's going to get even stronger with the benefit of the solvency review, I think. Pro forma, you're north of 230%.
Mm-hmm.
Admittedly, that's not necessarily deployable cash capital, but it's a very strong balance sheet. What can you do with that balance sheet that you're not already doing to the benefit of investors, do you think?
First, going back to our AI debate, a very strong balance sheet is critical in general in order to gain market share. Also, in terms of brand perception and the quality of the solutions we can offer to our customers. It's true in the commercial business, but more and more I believe it's becoming extremely important also on the retail side. By the way, if you take the view that many retail customers are gonna ask more questions before getting an insurance product, having a strong balance sheet is gonna be one of the elements that I believe LLMs will take into consideration when they ultimately make a recommendation. That was not the case on price comparison websites. Again, it's a competitive strength for us. That's the first way we look at it. Second way, we want to be above 200%.
We have been extremely clear on that. Everything we have above that is providing us with additional flexibility. You should look at it as a global picture. Ultimately, for example, that can give us flexibility on the leverage side. We'll see down the road. But it's, again, good news to have a strong balance sheet. We will maintain the discipline, and we have also been very clear on, I think, the quality of our distribution policy. You said it, but I want to reiterate that. Ultimately, this also strong balance sheet is helping us grow organically in each and every of our local markets. Today, we have also sufficient resources to do that while having the good distribution policy we have.
If I were to play devil's advocate.
Yeah.
One question, Mark, I've sort of toyed with in my own mind is your returns to shareholders. You've got a 75% payout ratio, the 60% plus 15 from the share buyback, which is, you know, consistent with what we're seeing elsewhere. But when I look at it from a free cash flow perspective and how much cash you're paying up to the holding company net of central costs and what have you, beyond what you're paying to shareholders, there isn't huge amounts left on an annual basis, so you're not building up a stock of cash at the company. Whereas if I were to do the same analysis for some of your competitors, they are generating more cash, which gives them a bit more flexibility potentially.
How are you thinking about that given the, you know, that there isn't much room for maneuver from a remittance perspective? Is there more to do from a remittance standpoint? Or how should we think about that?
Well, first, I would say that we are happy with the current level of remittance from our local operations. We are very much in line with the targets we gave you in terms of free cash flows, and we are happy with the 80% on average we are having year- after- year. It's again helping us to keep the necessary resources at local level to fuel organic growth. Ultimately, that's what we need in terms of free cash flow in order to be able to meet the distribution policy we committed to. We are very happy with this situation.
You're right, we are not creating huge amount of cash at holding level, and I believe it's good news because it's helping us maintain the right discipline in terms of distribution policy, and it has been on purpose designed that way, precisely because we wanted to give you a clear message that our intent is not to have, I don't know what kind of available cash at holding level, where then you would start wondering, are they going to do quote-unquote crazy things with that. You have your answer.
Very clear. Thank you very much. One area of the business that we've not touched on yet is the life side of things. I think it's fair to say that, you know, in the past, there have been some pressures on the traditional savings part of the business there. You know, more recently, new business momentum is definitely starting to gain some traction. How should we think about that in terms of translating into CSM growth and ultimately earnings growth for the life business? Because, you know, on an underlying basis, again, your life CSM growth looks slightly lower than your peers. How should we expect to see the momentum accelerating from here?
I would start by making a cautionary statement in front of you for all of us. I'm not entirely convinced that IFRS 17 drastically improved our ability to compare with each other. I tend to even believe the opposite, if I'm honest. This being said, when I look at the life business, in the end, it's, in my opinion, not that complicated. You need to look at the level of reserves, and you need to look at the level of margins. Year- after- year, are we able to at least maintain margins, if not improve them, and are we able to grow our results? That's exactly what we did in 2025. That's what we will continue to do in 2026. That's what we will continue to do going forward. Work on the margins, and you see that in our indicators. Work on the level of reserves.
The name of the game there is to make sure that net flows are growing. We have now really built the right franchise to go back to net flows growth. It's true that over the past five years, we had our homework to do in order to reposition our life and savings business. Now we are in the right position in order to grow. Net flows have grown in 2025. We will see that also in 2026. How? Stronger commercial momentum in the geographies where we have strong distribution networks in order to distribute our life and savings solutions, mostly France, Switzerland, Belgium, Italy, Japan, and Hong Kong. Also better persistency, because our book is now positioned exactly where we want it to be with good level of margins. That's the way I look at the life business.
If we keep working in this direction, you will see improvement in the CSM, you will see growth in earnings. Just one element that I believe is important, let's not forget that a lot of the elements that you see on the CSM are extremely dependent on the market you're operating in. For example, NBCSM, you look at it on the face value. For AXA, it was 3% growth. If you normalize that for interest rate, we in particular have a significant business, for example, in Japan. If you normalize that for interest rate, it's 5% growth in 2025. That's the organic potential, the real operational performance of the business. 5% NBCSM growth, I believe, is good, it's healthy, and again, it's in line with predictable, consistent earnings growth year- after- year, also on this business. I see the situation as being healthy.
We need to keep improving quality of the solutions and distribution network depths. On that front, quality of the solutions. The good news for AXA is that we are now a pure insurance player. We no longer have the question, "Oh, for the quality of my asset management solution, do I need to take into account other constraints?" Now, we sold our asset manager. We have a strong partner that is helping us improve the quality of the solution on asset classes where we were not best in class. I think about equity, I think about ETF. But we have also more freedom to work with competitive third-party asset managers. In the end, what I want is the best solutions for our customers. We have more freedom on the innovation side, and that will help us drive better proposition for our customers.
On top of that, on the distribution side, I believe that where we need to improve is third-party distribution. On life, in particular, we have high-quality proprietary distribution, but we can do more on third-party IFAs, brokers across the board, and we are investing a lot on the quality of our platform for those distributors.
Excellent. Thank you very much. I'm very conscious of time, so maybe let's open it up to the audience, see if there's any questions. Please, raise your hand and wait for a mic. Question from Mark.
Hi, it's Mark from BDL. Just, I noticed that at the beginning you were talking about efficiency and you had referenced the head office cost, group central cost. Can you just maybe expand on what you're thinking there in terms of the potential?
Well, we will on September 21st, but I do see as one of the leverage we have indeed for the future, and it's a healthy question to ask ourselves. We can always generate more productivity also with our corporate center. I see that as an opportunity.
Okay. With that, I think we can end it there. Guillaume, thank you very much.
Thank you all very much.
A very interesting conversation.
Thank you.