AXA SA (CS) Earnings Call Transcript & Summary

Welcome to Paris, and welcome to AXA's Investor Day 2020. It's great to have you with us. This morning, we're going to cover 6 topics, which we know are of importance to our investors. First, we're going to review the achievements of our last 5-year plan, Ambition 2020, and introduce you to our new plan, Driving Progress 2023. Second, we're going to take a look at our operations in France and Europe, which represent 60% of the earnings of the group. Then we'll talk to you about the progress we're making on improving the underwriting performance of AXA XL, then a short film on our leadership fighting climate change. And that will be followed up by a presentation on a topic, which, again, we know is important to our investors, cash and capital. We'll finish the morning by sharing with you our earnings targets for the next 3 years. Of course, at the end of the morning, we'll have a Q&A session. We'll be happy to answer your questions. So to kick things off this morning with a presentation on our strategic priorities. I'm very happy to hand over to AXA's Group CEO, Thomas Buberl.

Thank you, Andrew. I'm very happy to have this important session with you today. I know we've all got used to digital means, but I have to tell you I do miss the face-to-face, and I really hope that we see each other soon again. Today, we want to present to you our Ambition 2023, Driving Progress. This ambition is very much in continuity with the vision that we have set out in 2016. And again, it will be focused on profitable growth and sustainable value creation. When we look back on Ambition 2020 and what we have delivered, we can clearly say that AXA has delivered across all strategic themes. The first theme was very much a shift of our profile towards technical risks. We've achieved a significant transition of our portfolio to technical risks. Today, over 80% of our profits come from technical risks, and you've all accompanied us through the very big movements of disposing of AXA Equitable, acquiring AXA Catlin and certainly doing many other transactions like the transformation of the Swiss Group Life book. Today, we are very much focused on lines of business that have a much higher growth. The second theme that we've been working on was how to make AXA simpler. A first action was at the time to reduce the regions, to reduce the corporate center and to bring decisions closer to the place where we are really close to the customer. Of empowerment and simplification, we have really, really reached that aim. And you've seen it in the Net Promoter Score numbers that have been going up over the plan. At the same time, we've also focused our geographical footprint on less geographies with higher scale. And we made another important acquisition, which was the 100% ownership of AXA Tianping in China, which makes us the biggest foreign insurer in China. On the third dimension, we've really progressed to scale innovative services. These were across many places, but, in particular, on health, very much in line with the logic, how do we change our business model from purely paying claims to really becoming a partner of our customers. Amongst the many examples that we have seen, one has really struck me during the COVID crisis, which was the fact that we have delivered over 500,000 [ Telex. ] When you look how this has translated into financial numbers, we can clearly say that Ambition 2020 has delivered a very strong financial performance. We've been growing in our preferred segments, Health & Protection and Commercial Lines. We've achieved stronger technical margins. We continued our discipline on expenses despite a very challenging context of low interest rates, but also of COVID this year. If you look at the 4 metrics that we've laid out at the very beginning of Ambition 2020, we've reached the objective earnings per share, 5%. We've achieved our objectives on the cash generation, EUR 25 billion over the planned period. We've achieved the objective on the return of equity, 16%. And all of this with a solvency that was very high at 180% and a very solid balance sheet. Where do we stand today? The group is clearly reshaped. It is more focused. It is simpler. And when you look at AXA today, there are 3 main attributes. Number one is the unique leadership in Europe. EUR 4 billion of underlying earnings are delivered by very strong positions in Europe and France. Almost everywhere, we have top 3 positions, very solid and a very good contributor to the last ambition but also to the next ambition. A second very important attribute is our global leadership in the Commercial line space. EUR 32 billion of P&C Commercial line revenues, the biggest global platform that there is in insurance, in a time we are facing a hardening price cycle in order to make sure that our underwriting becomes even more profitable. And then thirdly, we've got some very promising growth avenues. Asia is one of them. We are 1 of the top 3 players in Asia and have very strong ambitions and clear strategies on growing the importance of health business in Asia. In the international and new markets, so Mexico, Egypt and so on, we are focusing very much on buying clinics and making sure that we are also going into the delivery of health services. And clearly, on asset management, where we have one of the most important and largest franchise in alternative assets and have become over time a great expert in ESG assets. COVID has certainly changed our environment. But I do believe that AXA is uniquely positioned in this new environment. What are those long term trends? And how do I believe can really benefit from this? Number one is, we clearly see an environment in which, on the one hand, interest rates remain low for longer, but this leads to price increases in technical risks. As I said earlier, being so focused today with over 80% of our portfolio on technical risk, it puts us into an excellent position to really take advantage of those price increases. The second topic is clearly around Protection & Health needs. We've seen that health risks are today the #1 risks and being so strong, both on the health side but also on the commercial side, with many, many customers as enterprises, we are uniquely positioned to take advantage of this trend. A third topic that we have seen arising a lot during that crisis was the desire of our customers for more simplicity and for faster interaction. We were all very glad that AXA, from very early on, invested in technology to make the interactions simpler, to make them more digital and to also integrate our agents into this digital journey. There is clearly a strong desire to go further. And then lastly, we've seen that solidarity and societal engagement becomes even more critical after this very difficult COVID experience. Companies do need to act in society and companies need to be an accelerator of societal change. AXA has been for very early on a leader in the climate transition. Climate transformation, green investments are becoming more and more important in a world where we want to build better. And therefore, we are uniquely positioned to take advantage of this and to really make it a strategic objective and core for our strategy to contribute to society. What is this new Ambition 2023, Driving Progress about? It focuses on 5 key strategic actions. The first one is we want to expand our Health & Protection franchise through services and across all geographies. We are today one of the largest player in this field. We have developed, as I've showed earlier, a lot of innovations over the last years, especially in health. And we want to achieve a growth in health over 5% on between now and 2023. Many, many services have been developed, over 90 of them in the entities, and there's a very strong commitment of each entity to make this even more of a reality to help our customers even more in a post-COVID environment. The second priority is around how can we simplify even more the customer experience, but at the same time, accelerate efficiencies. This is particularly important for France and Europe. Simplifying customer experience means that, on the one hand, we need to clarify even more our proposition, our products. And certainly, the learnings from COVID-19 really give us ample opportunity to do that. It means also that we need to continue redesigning and automating our processes that interact with the customer, but it also means that we need to fully embrace and fully integrate our agent and distributors into this customer journey. At the same time, we have set ourselves a clear ambition to accelerate efficiency and reduce our cost base between now and 2023 by EUR 500 million. The third strategic action is clearly around strengthening our underwriting performance. In many areas, we are already at a very high level, and we want to get better. But the one place where we really need to strengthen the underwriting performance and where we are benefiting from a hardening market is at AXA XL. And you will see in a few minutes how Scott Gunter will lay out to you with a very clear strategic action, with very disciplined underwriting, how he will reach a much better underwriting performance and how we reach the EUR 1.2 billion target that we have set ourselves for AXA XL. Fourthly, we want to sustain our leadership position in the climate transition. It is absolutely critical for me that society engagement has to be at the core of our strategy. We are very unique with that, but we've learned from the past that being truly aligned in the purpose of our company and our actions within and towards society is something absolutely critical. The continued leadership will be around an ambition that is how can we reduce the carbon footprint of our general account assets by 20%. This means that we all need to work on this and that we all need to put in the same direction from an investment perspective, but also from an underwriting perspective to make this a reality. And fifthly, and very importantly, we need to continue to grow our cash flows across the entire group. Making sure that we upstream enough and significant cash flow to service an attractive dividend means that we need to certainly continue on increasing the cash remittance of each of the local entities. We need to continue to reduce our in-force exposure on the general account, but we also need to make sure that we continue to reduce the footprint and focus AXA Group going forward even more. In the next few sections, you will certainly see more detail of that for Europe and France and for XL. Back to Andrew.

Thank you, Thomas. You might have noticed that the 5 strategic actions that Thomas just explained are precisely the topics which we're covering together this morning. The first 2 on growing our Health & Protection business, on customer experience and efficiency are topics which are important across the group, but particularly in our operations in France and Europe. Our CEO for AXA France, Jacques de Peretti; and our CEO for AXA in Europe, Antimo Perretta, will take you through those elements for France and Europe now.

Thanks, Andrew. Good morning, everyone. I am extremely happy to have the opportunity together with Antimo to walk you through our ambition for AXA France and AXA Europe in the next couple of years. Some of you might wonder why we decided to have a joint presentation today, whereas we usually discuss separately France and Europe. The reason is simple. Antimo and I have a lot in common. It's not only because we are both seasoned insurance executive with strong on the field and AXA experience, but mainly it's because we both operate together in the second-largest insurance market worldwide and in mature geographies that have similar pattern. We both ran entities that have played a pivotal role in AXA growth since 2015, that have proved very resilient despite turbulent times and that have developed highly differentiated assets and led solid foundation for future growth. And if we look towards the future, we see more similarities. We ran entities that share the same strategic priorities, namely: growth in health, simplified customer experience and cost leadership. And last but not least, we firmly believe that France and Europe will remain significant contributors to our new plans, Driving Progress 2023. That is what we are going to demonstrate during this presentation. And I will now hand over to Antimo for the first proof points.

Thank you, Jacques, and good morning to all of you also from my side. Before presenting our future strategy, I would like to take a moment to show you what makes AXA truly distinctive across France and Europe today. First, we have a unique proposition across our geographies. These are highly profitable insurance markets with significant potential to grow in our preferred segments. As you all know, these are large and sophisticated markets. And despite the competition, we strengthened and consolidated through the years, reinforcing our top-tier position. We are in top 3 in most of our markets, with EUR 61 billion revenues, thanks to our dedication to innovate and to satisfy our customers. We are proud to serve and protect every day a growing and loyal customer base of 38 million clients. This is a strong base for cross-selling across our preferred segments. We developed a leading multiaccess distribution franchise of over 15,000 tied agents with a very strong local footprint, 30,000 brokers and direct channels. The second driver of our distinctiveness is our capacity to deliver on our business transformation and the resulting well-diversified portfolio. This is key to reduce volatility and announce resilience. We count on a strong quality of our portfolio with a reinforced presence on our preferred segments contributing now to 71% of our revenues. We have also diversified on our clear strategy to reduce our general account portfolio with our Life in-force transformation programs. This is the result of our strategy to shift from financial to technical risk. And finally, despite adverse market conditions, we have delivered consistent earnings over the past years. We continue executing our strategy with rigor and discipline, leveraging from higher customer centricity, improved technical skills and lean insurance operations. And now Jacques will deep dive into how we performed across our main lines of business.

As Antimo just mentioned, I will now provide more details on how we transform France and Europe into a highly profitable growth engine with strong potential for the future. First, I would like to discuss about P&C, which accounts for more than half of our European earnings. Our combined ratio improved by about 3 points during the past strategic cycle and landed at 92.4% in 2019, which place us among the most profitable players in France and in Europe. How did we get there? Three reasons: first, a step change in underwriting by upscaling our agents, sophisticating our pricing and using more data. Second, insurance procurement excellence. We saved roughly 5% of our retail claim cost in 2019, notably by leveraging our European purchasing center for motor paints and parts. And third, cost efficiencies on the back of digitalization, offshoring and nearshoring initiatives. Going forward, we believe we still have plenty of room for improvement across these 3 dimensions. And we are confident that this will decrease our combined ratio by 1 point by 2023. Now, let's speak about Life & Savings. Over the years, we have built a resident Life & Savings business with an attractive earning mix that relies for 2/3 on protection annual savings and a sound G/A business with strong investment margin in the range of 80 bps and well-managed guarantees. This is the outcome of our disciplined, long-term-oriented strategy, which enable us to anticipate changing market condition and laid a solid foundation for the future. Our strategy revolves around 3 main pillars: discipline, innovation and in-force initiatives. First pillar, strict discipline on G/A. We reduced guarantees. We gradually decreased our profit sharing rate. And more recently, we introduced new sales commission with minimum UL rates, for example, 30% to 40% of new business in France. The second pillar is innovation. Since the early thousands, we have shaped the market by pioneering in UL product innovation, such successive launch of structured real estate, private equity and more recently, infrastructure funds. This successive innovation enabled us to encourage client asset diversification and accelerate the shift towards capital-light product. But they also enable us to meet customers' demand for attractive returns in the context of low interest rates. The last pillar is bold in-force initiatives. For example, the transformation of our Life business in Switzerland and Belgium, which enable us to derisk our balance sheet and upstream cash. In the coming years, we will stick to our strategy of disciplined, innovation and in-force initiative. And for example, we will further reduce our G/A exposure while increasing the UL rate on new sales by 10 points from circa 40% today to 50% in 2023. Now, after P&C and Life & Saving, let's turn our attention to health. Health has been and will remain the cornerstone of our strategy, as Thomas mentioned in his introduction. And the recent COVID-19 crisis highlight how critical it is when it comes to health to offer not only a comprehensive coverage, but also innovative solution beyond traditional insurance. This is exactly what we have done with the development of innovative value-added services. Our most emblematic health service is probably telemedicine, which we offer now to almost 6 million beneficiaries throughout Europe. Another emblematic development is a vertical expansion strategy that we implemented across multiple countries, such as Italy and Spain. These game-changing initiatives enable us to deliver a solid 6% per annum top line growth between 2016 and 2019. And we are confident, Antimo and I, that we will be able to maintain this strong growth momentum in health in the coming years. Let's now move to the next page to find out how. To expand in health, we will further reinforce our service offering. And Antimo will give you more details about what -- that in a moment. But first, let me highlight that we will also scale up our employee benefits franchise across all the region of the group. Why are we so keen to accelerate in this segment? First, because employee benefit is a large, growing, high return on equity market at the intersection of our top priority segments. Second, because it is a fragmented market without any true global leader. And third, because we believe that when operated at scale with the right expertise and tech solutions, it offers significant opportunities for profitable growth. So what are our competitive advantages today? We have built a sizable and profitable EB business totaling EUR 13 billion premium with stronghold in 4 countries. But most importantly, we have developed a truly differentiating model relying on 3 key assets. What are they? First, a strong local footprint with our local entities. And that is essential to get client access to develop a deep understanding of local market dynamics and to operate local servicing capabilities. Second, fast-growing international businesses, which provide sophisticated insurance and reinsurance solutions to multinational companies, expats, local affluence as well as local public/private health carriers. And last but not least, a proprietary insurtech called EB Partners, that has developed successful digital EB solution, which are now already used in 9 countries of AXA. Going forward, our ambition is to take advantage of this asset and deliver high single-digit growth through selective organic growth, the rollout of EB Partner's tool in new countries and the setup of global life center of excellence to support local teams. Last but not least, we will reinforce our relationship with our customers by offering differentiating services. This is now what Antimo will present.

As Jacques just mentioned, we will make a step change for our private health customers through services. The current health market is made of fragmented customer journeys and low personalization. Over the past years, we have focused on addressing these issues through innovative proposition in our 2 health stronghold, the U.K. and Germany. In these markets, we were extremely successful in connecting services, leveraging data and offering an integrated proposition. This enabled us to gain a deeper understanding of customer needs and provide better quality of care to millions of customers. Therefore, as next step, we will scale up this concept across Europe to establish AXA as the leader in the health services. We will do so with the ambition to reposition ourselves as the orchestrator of a health ecosystem. Concretely, we are building a pan-European health care ecosystem by connecting innovative services. We are using technology and data to improve customer propositions. In parallel, we will leverage AXA global footprint to scale up best practice across the group. For example, Emma, our integrated digital well-being and prevention platform in Asia. One other example is our strategy of expanding alone the health value chain by accruing clinics, diagnostic centers across Italy, Spain, Mexico and Egypt. To conclude with this evolution, we aim to do 2 things. First, to improve customer experience and quality of care for our 10 million health customer across Europe and France. Second, to drive claims cost reduction, which includes prevention. This action will significantly contribute to drive health revenue growth of more than 5% in our new plan. That concludes the deep dive of health. Jacques will now discuss you our second top strategic priority, customer experience.

As Thomas said earlier today, one plan of our strategy is to simplify customer experience to increase satisfaction, as Antimo just said. That is notably true in P&C Commercial lines. P&C Commercial line is a profitable gross engine for both France and Europe, with a revenue growth of 4% per annum in the recent years and a combined ratio, which now stands at 94%. So our ambition is to expand our leading market position, especially within the SME to further improve our profitability. We will achieve these objectives by taking advantage of 3 pillars. First, our dual distribution model, which relies on a growing number of expert agent with 650 additional SME specialized agents across France and Europe and brokers with long-standing relationship with AXA. Second, our differentiating payer to partner services, such as cyber and on-site prevention. Third, significant commercial synergies with AXA XL. But most importantly, we will reach our goal by augmenting our Commercial lines agents with innovative digital advisory and underwriting tools, which will deliver a superior customer experience. OSE, that is mentioned on the left-hand side of the page. It's a very concrete and relevant example to illustrate that point. What is it exactly? A digital underwriting tool, unique on the French market, representing a EUR 50 million investment and jointly developed with agents and underwriters. It leverages internal and external data, such as license plate numbers, corporate IDs or satellite images. As such, the coat and buy process is on average 40% less time-consuming. It gives agents and brokers the underwriting authority below certain threshold while fostering technical excellence. And at last, it facilitates cross-selling initiative as it is built on a unique platform. As you can see, it is a very powerful tool, which already strongly contribute to our Net Promoter Score, which is 10 points above the market. During the next plan, we will roll out it from 55% of our P&C commercial line new business in France to 75% by 2023. And obviously, we launched similar initiatives across Europe. For instance, the rollout of AXA Advice Antimo spoke already about. That gives us a flavor of what we are doing to boost growth while improving customer satisfaction in P&C Commercial line. Antimo now will tell you what we are doing on retail.

Thank you, Jacques. Transformation is key to our strategy across all business lines in France and Europe. Over the past years, we have focused our transformation journey on all processes and offerings to become more customer-centric. We redesigned our customer journey across the entire value chain. At every step of the way, we focused our thinking on the customer first and foremost. We made 4 important changes: first, we introduced digital and advisory tool; second, we simplified policy wordings and communications; third, we designed simpler and more intuitive propositions; and lastly, we offered added value services beyond the traditional insurance contract. The end result is that it boosted customer satisfaction and fund financial results. Our NPS is now at or about market average across most of our businesses. Importantly, we have moved from negative to positive net new customers. Customer satisfaction is a win-win and pays off. All of these improvements are driving our growth ahead of competition, especially in our preferred segments. Going forward, we will accelerate our transformation by upgrading to a unique digital advisory and distribution powerhouse. We will do this in 2 ways: first, a seamless customer experience. We will redesign 80% of our customers and enable them digitally. Second, a best-in-class, digital-enabled advisory. We will keep 100% of our adviser with digital tools like AXA Advice, which allows a professional 360 degrees customer advisory. As we have seen, best customer experience and satisfaction will be key drivers of success in the future. Simplicity and productivity are 2 sides of the same coin. Over the past years, we have deeply transformed our business by leveraging technology and data. Across Europe and France, we have implemented major changes in our organization structure, intelligent process automation and IT modernization. In a nutshell, we have reduced our overall cost base by buying increasing investments in our transformation. Over the next 3 years, we will continue to accelerate our transformation and efficiency measure. By 2023, we will reduce our cost base, France and Europe, at net EUR 300 million. We will do so by focusing on 3 layers. First, we will modernize our IT infrastructure. Over 90% of applications will be on the cloud. Today, we are already about 50% while competitor are at 10% or below. We will decommission an additional 30% of our legacy system by 2023. Second, we will take automation and digitalization on the next level. We will increase straight-through processing. In motor, we will reach between 70% and 90%. We will also simplify our produce offering, reducing by 40% across Europe, with some countries like Spain down 80%. Third, we will simplify the way we operate and optimize our real estate. In AXA France, we will simplify support functions by reducing layer, redefining roles and simplifying responsibilities. In addition, we will consolidate and transform our offices, the suiting in a real estate consolidation of minus 25% by 2023. Thanks to this strategy, we will be a leaner, simpler and more agile company by better serving our customers. Now, let's move to the conclusion. In conclusion, AXA has a unique position across its key markets in France and Europe. During Ambition 2020, we deeply transformed our businesses. Going forward, we will accelerate our transformation, delivering along 3 axes: first, growing in preferred segments; second, increasing technical excellence; and lastly, innovating to become more nimble and more productive. In the face of growing competition and economic uncertainties, we remain confident, Jacques and I and all the team, to execute of our plan, thanks to our strong leadership, talents and our culture. And now I hand over to you, Andrew.

Well, thank you very much, Antimo and Jacques. Very good. We're now going to pass to the third of the strategic actions of our new plan. And that is about improving our underwriting performance, notably at AXA XL. And to join us and to take you through some of the detail of that, we're going to be joined now by the CEO of AXA XL, Scott Gunter. Now obviously, he couldn't be with us physically in Paris this morning, but he is going to join us from the U.S. Hi, Scott. Can you hear me?

Hi, Andrew. Perfectly well. Thank you.

Great. Over to you.

Good morning, everyone. Today, we are going to talk about 3 key topics: the first is to give you an understanding of the actions we have taken to deliver on our earnings target in 2021; the second is to give you some color on how we are driving improved underwriting performance across the portfolio; and third, the steps we are taking to reduce the volatility of our earnings. Before I get into the changes that AXA XL has undergone since I joined in February, I want to give you some of my background and why I joined AXA XL. I'm an underwriter by training, having spent over 30 years at Chubb in a variety of roles, predominantly in the commercial insurance side of the business, including Global Chief Underwriting Officer for Chubb's commercial insurance. I have learned a lot over those years, but the one lesson that really stuck with me was the importance of underwriting discipline in order to achieve your earnings goal. After joining AXA XL, I've been impressed by the strength of the organization, including the depth of our technical talent, how focused we are on our clients and brokers and our broad portfolio of products and services. We also have an impressive geographic footprint with colleagues in 30 countries and through our network of partners, we can issue policies in over 200 countries. We can service our clients as they expand around the world. Combine this with the financial strength of AXA Group, and we have a compelling value proposition for our clients and brokers. While we have tremendous potential in our organization, I also found that we had some work to do to achieve the kind of results we are capable of achieving. One of the first actions I took was a review of all of our business globally where we are not achieving an underwriting profit. I will provide more detail on the outcome of that work in later slides, but it became clear that we needed to make changes to how we are organized. We had to find a way to enhance accountability, improve our underwriting discipline and take steps to reduce the volatility of our earnings. Before we get into how we implemented the necessary changes, I would like to give you a brief snapshot of our business on the next slide. With the addition of XL, AXA Group became the largest property/casualty commercial lines insurer. As you know, with XL and AXA coming together, we have the capability to ensure the smallest commercial clients to the largest companies in the world. The middle chart is a breakdown of AXA XL. Our portfolio was almost EUR 20 billion with insurance making up 75% and reinsurance 25%. From the doughnut-looking chart, you can see that our insurance business is well balanced between specialty, property, casualty and financial lines. The chart on the right-hand side shows our insurance portfolio in terms of average combined ratio and the combined ratio volatility versus our peers. As you can see, AXA XL's performance is in the middle of the pack relative to our peers. The issues that have impacted AXA XL in recent years, such as the prolonged soft market, social inflation and the escalating natural catastrophes have impacted us and the industry as a whole. While being the largest is important, our ambition is also to be one of the best performing. The steps we have taken in 2020, and will be taking in 2021, will support our movement towards a lower and more sustainable combined ratio. I am confident in the ability of our organization to execute as we strive to become the best-in-class commercial lines insurer in the marketplace. One of the key elements in improving our results is increasing pricing. Let's now take a look at what is happening in the property/casualty marketplace. The graph on the left shows a period of prolonged rate softening from around 2012 into early 2018. From there, we have seen pricing across more and more products turn positive. The first part of the U.S. market to see pricing increases was the commercial automobile business, followed by property due to increased catastrophe losses. Following property, the significant increase in loss trend as a result of social inflation pushed rates up in casualty and financial lines. This momentum was accelerated in 2020 by COVID-19 losses and economic uncertainty. We are now seeing positive rate change in virtually all geographies and products. This global firming market allows us to do 3 things: one, achieve better pricing on both new and renewal business; two, we can adjust terms and conditions to pre-soft market levels; and three, we can take advantage of opportunities where we had previously declined the business due to inadequate pricing or unattractive terms and conditions. As you see on the right-hand graph, the rate change we are achieving in AXA XL insurance is in line or ahead of the industry benchmark. We believe this is a result of our strong value proposition, including our global capabilities, the breadth of our client relationships across multiple products and a flight to quality as clients value the financial strength and stability of AXA XL. Improving pricing is just part of the story. But now let's move on to talk about what steps we are taking at AXA XL to improve our underwriting performance. Let's unpack this slide a little bit. On the upper left-hand corner, we are pleased that the integration of AXA XL is now complete. We have successfully brought together 3 companies, XL Catlin, AXA Corporate Solutions and AXA Art. Our ability to achieve our revenue synergies target is evidence of the support we see in the market for AXA XL as well as tangible proof on how well the teams across both AXA XL and the AXA Group are collaborating, providing more clients the breadth of products and services we now have to offer. The second item is that we continue to foster our depth of underwriting talent, and we're working hard to keep focused on our clients and brokers especially during these challenging times. The third item is our global footprint. As mentioned earlier, we have operations in over 30 countries, and we have the ability to service our clients through local partnerships in over 200 countries. We are where our clients are. The bottom 3 items of the slide show the changes we are making. Building on the strengths of AXA XL as well as addressing our challenges, we've been focused on 3 key priorities since I joined back in February. I will use these 3 priorities as the basis for the rest of the presentation. First, we are going from a complex, somewhat opaque organizational structure to one that has clear profit and loss accountability. Second, we're moving from a primary focus on top line growth to a focus first on underwriting profit and then on top line growth. Third, we are reducing earnings volatility through a more consistent approach to the size of our risk limits as well as purposefully reducing natural catastrophe exposure through underwriting actions and increasing ceded reinsurance coverage. The next slide gives some additional detail on our new organization. To establish clear accountability, we implemented a new operating model where we created 4 business units, led by a Chief Executive focused on the execution of our tactics and goals in our countries around the world. These business units are supported by a global underwriting office and other global functions. These 4 business units are the Americas, North, South and Central America, including Bermuda; Asia and Europe; U.K. and our Lloyd's platform; and Global Reinsurance. The change in leadership and the new organizational structure allows for 3 key strategic focuses. One, the business unit CEOs now have a clear profit and loss accountability and the responsibility to achieve our goals around client and broker service as well as underlying earnings targets in their areas of responsibility. Two, we have moved more accountability closer to the client, enabling the tailoring of go-to-market offerings based on regional nuances and the client requirements supported by the global functions. And three, we significantly enhanced our focus on underwriting discipline and consistency through the creation of a global underwriting office. It was critical that we move quickly in developing and rolling out our new way of working. We launched the new structure in April, and we move quickly to get the right people in the right roles. During these times, morale is of the concern. As a result, we conduct regular internal colleague surveys. Despite making significant changes to our organization during a prolonged period of remote working, our most recent internal survey indicates that our colleagues are responding well to our underwriting and service strategies as well as the organizational structure. Despite the challenge of working in a remote environment, our colleagues are getting more and more settled into the new way of working. Through all of the changes, the team has continued to keep their focus external, providing service and solutions that meet our clients' needs. Now let's take a look at our second priority on the next slide, underwriting discipline. As I mentioned in the introduction, one of the first actions I did when I joined the company was to do a deep dive into the business, focusing on those products that were running over 96% combined ratio. The goal of these deep dives was to both understand the business and to figure out what actions we could take very quickly to start to turn around the results. This same approach applied to our planning process in 2021. We started our '21 planning with a more granular approach to establishing the base loss ratios. We then determine the action plans to either grow the product where we're achieving sufficient margin or improve the profitability where we were underperforming through higher rates, lower retention and/or less new business. Our planning, our insurance and reinsurance business, we've been very firm that only these products with strong underwriting performance have earned the right to grow. Result is what you see on this slide, with an average combined ratio expected at approximately 96% in 2021 for our total portfolio with no products above 100% and close to 70% of our products at or below 96%. As we move from 2021 to 2023, we expect to see more products and geographies earn the right to grow due to improved rate and the underwriting actions we have taken. On the next slide, let's take a look at some of the large product lines where I will give you some additional color of the work we've undertaken in 2020. What we have on this slide is examples of the key products that we are fixing, products that we are doing okay, but we still need some additional attention, and a couple of products where we're achieving appropriate underwriting margin and we are looking to grow in 2021. Let's look at the first 2 products on the list, International Financial Lines and North American excess casualty. As you have heard from us, and I'm sure others in the industry, we have seen poor underwriting results in both these areas in the last few years. While both have challenges, our approach to fixing them has been different. For excess casualty from 2016 onwards, the industry saw a significant jump in the loss environment in the U.S., referred to as social inflation. Social inflation is the term used to describe the rising cost of insurance claims from the increasing litigious environment, largely in the U.S., with widening definitions of liability and large compensation awards. This is particularly impacted policies for the larger corporate clients where limits are typically larger. Due to the benign loss trends in 2012 and 2016, we, like many others, were deploying large limits in excess casualty, and it now appears they were inadequately priced. Starting late 2019, but really ramping up this year, we implemented 3 key underwriting actions to fix our North American excess casualty business. These are: one, we have reduced the capacity deployed per risk, limiting our maximum loss per risk. As a result, our average net limit is now 30% lower than it was this time last year. Two, we have increased our attachment points by 15% versus last year. And three, we've been able to achieve significant rate increases particularly in 2020, where we're seeing close to 80% price increase year-to-date on this business. As a result of these actions, for 2021, we expect excess casualty will be back into profitability with a combined ratio below 96%. We have deep underwriting talent in our North American excess team. And we feel confident in our ability to manage this business line profitably in future years. Compare this strategy to our approach on our International Financial Lines business. International Financial Lines has a number of products within it, with the main ones being management liability and financial institutions business. While we write this business out of many countries in the U.K., Europe and Asia, it is our business written out of London that was driving the poor performance. While it is never our preferred action, we took the very difficult decision to pull out of the management liability and financial institutions business in London earlier this year. As a result, in 2021, we are reducing the gross written premium of our International Financial Lines portfolio by approximately 50%, excluding the impact of rate. As we're now renew significant amounts of our business, retention ratios will be under 60%, which is well below average for our insurance portfolio, which runs typically in the low to mid-80s. For the rest of the International Financial Lines business, we'll continue to write this business in Europe and Asia, and we are currently seeing strong price increases of over 20% year-to-date and opportunities for tighter terms and conditions, including reduced limits and higher client retentions. We believe our actions will return International Financial Lines business to profitability in 2021 and to below 96% in 2022. Let's take a look at some of the other products on this slide. There are also opportunities to grow our business in 2021. For a U.S. professional, mainly U.S. public D&O business, we have a fantastic team in Hartford, Connecticut who's very well-regarded by both our brokers and our clients. This team continues to do an excellent job in risk selection, line size management and pricing, and we expect to continue to grow, albeit at a slower rate than 2020, well into 2021. Our international casualty book is comprised of large corporate accounts and upper-middle market clients in Europe, United Kingdom and Asia and Australia. This business was running a bit of a temperature last couple of years. But with careful capacity management, pricing increases and tightening terms and conditions, we expect to be profitable and to grow this business selectively in 2021. In terms of reinsurance, not surprisingly, our property catastrophe portfolio has taken a bit of a beating the last few years. As a result, we have specifically targeted reductions in our exposures, I'll speak more to that in a later slide and increased pricing, both in the U.S. and internationally, with areas like Florida and Japan seeing better than 20% increases on wind renewals. We will continue to carefully manage our net catastrophe exposure through the use of retrocessional reinsurance, including third-party and partnership capital. Going forward, we will continue to leverage our strong data and analytic capabilities to monitor exposures and deploy capacity on a best-priced opportunities. Finally, on U.S. middle-market casualty reinsurance, we've seen this business perform better than the excess casualty on larger accounts. In reinsurance, we have an experienced stable underwriting team. We are seeing opportunities for selective, deliberate growth where pricing meets our requirements. Generally, we are focused on increasing shares of existing clients. We have strong established relationships rather than adding new cedents. As with property cat, we continue to actively manage our limit deployment to monitor accumulations. Now, let's take a look at our third priority, reducing our earnings volatility. Our third priority for the year has been to take concrete steps to reduce the earnings volatility within our portfolio. We've increased on 2 key drivers of the volatility of our business. The net limits we retain per risk and how much natural catastrophe exposure we retain. In terms of reducing our net retentions, starting shortly after I arrived, we took a review of our net limits exposed. Based on my experience, in many cases, our debt limits were just too high. As a result, we put in place an initiative to reduce our maximum net retention on new and renewal business to EUR 25 million. This was done through buying reinsurance as well as, in some cases, reducing the gross limits we deploy. Reducing net limits has 2 main benefits. One, we limit the size of an individual loss, and we prevent a single claim from disproportionately impacting a portfolio or our earnings. And two, we can deploy our capital across a wider base of risks. In 2020, we have already reduced the overall count of policies with over 25 million net retention by 40%, with notable areas for reduction being North American exit casualty, financial lines, construction and specialty. This will be an ongoing effort through the next couple of years as we identify the most appropriate and economically viable strategy for each product. It is important that we maintain relevance to our clients who choose us for our ability to offer solutions across a number of product areas. The second step to reduce volatility is our exposure to natural catastrophes. This year, we took steps to reduce the volatility from our nat cat exposure across key peril regions such as Japan earthquake and North America windstorm in Florida and the Gulf of Mexico. We did this work across both our insurance and reinsurance portfolios. As a result, we saw a 20% reduction in aggregate nat cat exposure from 2019, and we expect to further reduce our nat cat exposure by an additional 5% to 10% through a combination of reduced gross exposure and reinsurance. On top of the work on exposure reduction, we've also taken a more prudent approach to our budgeted cat loss load for 2021. On a gross earned premium basis, we've increased our planned cat loss load from 4% in 2020 to 6% in 2021. We set the planned cat loss load at this level as we reviewed the most recent catastrophe loss activity, and we decide to lean into our more recent experience. This helps us reduce volatility of earnings as the higher cat loss estimate in our 2021 plan reduces the likelihood of exceeding our nat cat estimates. On that note, let's take a look at our 2021 underlying earnings plan. To begin with, for 2020, we expect normalized underlying earnings ex COVID, ex U.S. rides with a 4% normalized cat load to be around EUR 1 billion. As many of you remember, the original underlying earnings plan was EUR 1.2 billion. However, in 2020, while our pricing increases are stronger than we expected, these have been overshadowed by the increases in our 2020 loss ratios, reflecting more recent loss trends. As a result, the net impact to underlying earnings is EUR 100 million lower than our expectation at the beginning of the year. In addition, investment income is coming in EUR 100 million lower than expected. This is primarily due to lower interest rates in the United States and lower-than-expected dividend distributions. Now, let's take a look at 2021. Our ability to achieve rate in excess of claims cost trends as well as additional underwriting actions that we've talked about earlier will bear fruit in 2021. We are expecting a significant EUR 500 million in additional earnings from underwriting and pricing when compared to normalized full year 2020. On the negative side, we do expect EUR 100 million headwind on investment income linked to the impact of the COVID-19 crisis on the economy and financial markets. Also, remember, we are moving to a more prudent approach in our cat loss budgeting, increasing the load in round numbers from 4% to 6%, resulting in an increase in our expected cat loss of approximately EUR 200 million. Overall, we are targeting EUR 1.2 billion in underlying earnings in 2021, including the higher cat load. Looking at it from a combined ratio perspective, this means an improvement from a normalized 98.3% in 2020 to an estimated 96% in 2021. However, this is not the end of the story. I wanted to finish with why I joined AXA XL and what I hope to achieve as the CEO. I was drawn to the challenge at AXA XL as I feel there's tremendous potential for such a diversified and well-established organization to be an even more important partner to our brokers and clients. I spent most of our time today outlining the changes we have made to our structure and the approach to the business. The fixed stage is absolutely critical to building the foundation of underwriting profit, but equally important, I believe we are building the foundations of a great property and casualty insurance company. We are moving towards unlocking the full potential of AXA XL. We will continue to capitalize on the favorable market conditions and achieve strong performance in 2021 and beyond. Thank you for your attention. And now let's head back to Andrew in Paris.

Thank you, Scott. And Scott will be joining us again in a few moments for the Q&A, and he'll be very happy to answer your questions. We now pass to the fourth of our strategic actions, and we believe that it's significant that 1 of AXA's 5 strategic actions for their new plan is around climate. And Thomas is going to join us now to explain why.

Being a global insurer also means being in the middle of society. Many of our decisions on the underwriting and the investment side do drive progress and drive societal development, in particular, on the climate side. Our vision of changing from being a payer to becoming a partner is not only important for our customers but also for society at large. We've been very active in the last couple of years, in particular, on the climate front and in social inclusion, to make sure that we contribute to society and that we show our solidarity. This is also very much linked to the reinforced purpose of AXA that we have launched this year, Act for Human Progress by protecting what matters. This is very inspiring to all of us and gives us a lot of energy to do better. Have a look in the film, what we have achieved and what we have done since 2015. [Presentation]

We'll now move to the fifth of our 5 strategic actions for the next 3 years and a topic which we know is of importance to you, cash and capital Thomas, back to you.

Thank you, Andrew. I will now address implications of our plan in terms of cash and capital, the fifth pillar of our strategy, focusing on the following key topics: first, our cash generation, which will support our ability to continue to pay attractive dividends to our shareholders; second, further actions to continue to reduce our exposure to GA savings; third, the ongoing simplification of the group's geographical footprint; and fourth, our clear capital management principles. Let's now go deeper into each of these priorities, starting with our organic cash upstream potential over the plan period and cash accumulation at holding level. Over the next 3 years, cash upstream from operating entities to the holding is expected to be strong at around EUR 14 billion. This should ensure a progressively growing dividend over the plan. Cash remittance in 2021 will naturally be lower due to COVID crisis, which will recover in the subsequent years given our profile shift, and it would lead to cash upstream between EUR 5 billion and EUR 6 billion per year in 2023. Cash upstream targets will also have a stronger weighting in the management compensation at group level and across all entities. This will ensure a strong alignment in optimizing cash resources and managing cash centralization at holding level. On a strategic basis, assuming EUR 11 billion of dividend to shareholders over the plan period, this means that we should have around EUR 3 billion of additional cash available. Of course, this is on top of the strong cash position at the end of 2020 expected to be above EUR 3 billion, as explained during the 9 months 2020 analyst call. So as you can see, even in this difficult economic context linked to COVID-19, we do continue to expect to generate significant cash resources, which should ensure attractive and sustainable dividends to our shareholders. Now let me tell you about the management actions that contribute to our strong cash generation. Over the past years, we have taken very bold in-force management actions across all markets to optimize our general account exposure, which led to a reduction of our general account reserves of EUR 27 billion. In addition to that, if you look year-after-year, we have been consistently managing our general account savings net flows to be negative. Since 2016, our general account savings net flows have been minus EUR 7 billion. As a result, over the period 2015 to 2019, we have reduced our general account reserves in France and Europe by 7% while our peers have increased their general account reserves by 12% over the same period and for the same scope. If you look at AXA's general account reserve now, 40% of it lies in France with an in-force guaranteed rate of just 0.5% and a 0% guaranteed rate on new business before commissions, and so effectively corresponding to a negative guarantee. We will not stop there on reducing our general account savings exposure. Over the next 3 years, we will continue to manage our general account savings net flow to be negative while extending our in-force actions to further reduce the general account exposure to reinsurance, buyouts, disposals, freeing up capital and resulting in higher cash remittance. Moving on now to another type of management action, streamlining our geographical presence. Since 2016, we have taken significant steps in the simplification of the group. We have closed 10 transactions during the period, the most prominent one being AXA Equitable Holdings, and have further ongoing disposals signed and awaiting completion. We will continue the simplification strategy over the next years and selling entities which have suboptimal scale, low return on equity or low cash remittance potential. This will generate cash and will bring the opportunity to better deploy these resources. Our intention is to offset the earnings dilution from these disposals over the planned period through share buybacks. Now what about our capital management principles? Well, you see, they are very clear. Going forward, we will target to hold a Solvency II capital at around 190%, which is a 50% buffer above the risk appetite limit of 140%. Our dividend payout range will be 55% to 65% of underlying earnings per share, which is broadly equivalent to the previous 50% to 60% of adjusted earnings per share. This change of reference comes from our intention to simplify our non-GAAP indicators. We are also reinforcing our strict financial discipline on the use of cash. We've already made significant change to our business mix and have reached the right scale. Going forward, acquisitions will be systematically evaluated against share buybacks, ensuring that any access at holdings is deployed in the best interest of our shareholders. Overall, there are 3 key points to retain. Firstly, you can see cash remittance will remain strong, increasing to between EUR 5 billion and EUR 6 billion per year by 2023, ensuring sustained dividend growth over the period, an additional accommodation of cash at holding of about EUR 3 billion. Secondly, we will continue to work on in-force actions to reduce our general account savings exposure and on the simplification of the group through disposals. And thirdly, we have set out clear capital management principles, and you can expect us to be very disciplined deploying cash in the best interest of our shareholders.

So thank you, Thomas. And now having covered all 5 of our strategic actions together this morning, we're going to move to a wrap-up. At the beginning of the wrap-up, Thomas is going to share with you our earnings targets for the next 3 years. Thomas, I hand straight back to you.

AXA Group has a very interesting and focused profile, and we are convinced that our strategy will lead to sustained earnings growth. France and Europe will continue to deliver strong performance with growth and strong technical margins as well as cost reductions. The turnaround of AXA XL will deliver higher earnings with EUR 1.2 billion already in 2021 from decisive underwriting actions and benefiting from stronger pricing. Asia, International and AXA Investment Managers are further growth avenues to support our medium- to long-term growth profile. All of this leads to a target range of 3% to 7% underlying earnings per share, average annual growth over period of 2020 to '23. It is ambitious given the challenging macro context, but realistic considering our strong track record in the past years in a declining interest rate environment and the quality of our assets. This target range is based on a starting point of 2020 earnings adjusted for COVID and natural catastrophes beyond the normalized level. So as I said, both ambitious and realistic. We're going deeper now into the levers of our target range of 3% to 7% underlying earnings per share growth. First, selective growth in our preferred segments, such as Health, Protection and P&C, Commercial lines as well as Unit-Linked across the group, which will contribute 2% growth. Second, continued focus on technical profitability and underwriting discipline, notably at AXA XL is expected to generate an additional 2% growth. Third, cost efficiency measures leading to an absolute reduction of our cost base of EUR 500 million over the next 3 years, which will help generate an additional 3% growth. As you are aware, interest rates across the world are at an all-time low and even negative in many countries. The prolonged low interest rates are a structural headwind for us. 2% negative growth are coming from a progressive reduction in the investment yields. You would recall that in Ambition 2020, the interest rates impact on underlying earnings per share was much higher. This lower interest rate sensitivity is coming from all the actions we have taken on our portfolio over the past 3 years. All of this means a 5% underlying earnings per share average annual growth rate in our base scenario. Of course, our earnings growth trajectory will also, by nature, be dependent on the speed of the macroeconomic recovery following the COVID-19 crisis. Depending on how quickly the economy recovers, we foresee an additional support or headwind of 2% in terms of earnings growth. The level of GDP may have some impact across most earnings levers but is likely to be most notable in our capacity to grow. As a result, our range of underlying earnings per share average growth of 3% to 7% over the period of 2020 to '23 is reflecting the earnings growth levers under management control, combined with the potential impacts from the macro environment. To conclude, I would like to reiterate our 5 strategic actions, which are truly aligned with the interest of all of our stakeholders and are at the root of our new purpose, Act for Human Progress by Protecting What Matters. First, we will continue to expand in Health & Protection. We will do this across all of our geographies and notably by getting closer to our customers through innovative new services. You saw very concrete examples of that this morning. We are also targeting annual top line growth of more than 5% in health over the coming 3 years. Secondly, we will continue to simplify customer experience by digitally enabling our advisory networks and to reduce costs by EUR 0.5 billion through simpler and automated processes, IT and organizational redesign as well as office space consolidation. I believe Jacques and Antimo brought clarity on these points. Our aim is to be simpler, smarter and faster. Third, we will strive to strengthen our underwriting performance further, most notably in P&C and at XL. Scott has gone through the various actions already underway and has given a very clear path to higher profitability already in 2021. We target to achieve a P&C combined ratio of 93% at group level by 2023. This would make us one of the leading P&C players in terms of profitability, most certainly in the Commercial line space. Fourth, we will maintain our leadership position in the fight against climate change and in shaping the transition towards a greener economy. You have seen our climate report. You have seen all of our actions. To continue with this strong line of engagement, we will intend to reduce our carbon footprint by 20% on all of our general account assets by 2025. And beyond climate, the whole group is engaged towards building a better society, more inclusive and sustainable. We are moving our engagement model to the next level. We have reaffirmed and clarified our purpose and have embarked all of our employees in this journey. I truly believe this sets us apart from our competitors. Finally, and probably one of the most important topics for all of you, we will continue to grow cash flows across the entire group from our operations, but also further enforce actions on our Life general account book to ensure attractive and sustainable dividends to our shareholders. We target EUR 5 billion to EUR 6 billion annual cash upstream to AXA SA by 2023. On top of this, you can certainly trust us to continue the simplification of our group, focusing on our core assets and to deploy excess cash resources in the best interest of our shareholders. So to wrap up with our 4 main financial targets of this new plan: first, underlying earnings per share growth of 3% to 7% from 2020 to 2023; second, an underlying earning return on equity target of 13% to 15%, which is equivalent to the previous plan's adjusted return on equity target of 14% to 16%; third, EUR 14 billion of cumulative cash to be upstreamed to the group over the next 3 years, more than sufficient to provide a sustained dividend growth to our shareholders; and finally, our Solvency II ratio, which we aim at maintaining at around 190%. Back to you, Andrew.

Thank you, Thomas. We're now going to move to the Q&A. All of our speakers will be here to take your questions, and Scott will be joining us again from the U.S. We'll also be joined by some other members of the management committee; our CFO, Etienne Bouas-Laurent; our CEO of Asia, Gordon Watson; our CEO of AXA IM, Marco Morelli; and our Chief Risk and Investment Officer, Alban de Mailly Nesle. [Operator Instructions] And so now, we'll be very happy to take your questions.

The first question from Mr. Farooq Hanif from Credit Suisse.

I just want to turn to Slide 32 on your cash. The EUR 14 billion of cash remittance and manageable actions is clearly far less than the EUR 5 billion to EUR 6 billion of cash remittance. So can you explain difference or the path to the EUR 5 billion to EUR 6 billion. And without the additional debt, clearly, this is implying dividend cover of only 1x, which is a bit lower, I think, than your historical levels. So could you also comment on why you think this is sufficient to protect the dividend? Second question is really for Scott. On -- you haven't commented on reserves in this time. So I was wondering if you can give an update on AXA XL, specifically the BCAP reserve, your reserving strength? And maybe a comment on how that looks in the context of the group. And lastly, also for Scott, could you talk a little bit about top line growth ambition or potential for AXA XL, given all of the factors that you talked about, including underwriting discipline, line sizing, being careful about growth in certain areas. If you could just give us some ambitions there, that would be helpful.

Thank you, Farooq, for your 3 questions. As suggested the first 2 are being answered by Etienne. So your first one was very much on the question around the EUR 14 billion cash remittance and how this is really linked to the EUR 5 billion to EUR 6 billion. Second question is also for Etienne around the reserve level at AXA XL. And then the third question for Scott on the top line ambition at AXA XL. Etienne, why don't you start with the first 2, and then we pass over to Scott.

Thank you, Thomas, and hello, Farooq. Thank you for your question. This EUR 14 billion of cash upstream, first of all, allow for a high level of dividend and progressive level of dividend over the next few years, which I think is an important information. The second thing is, as you know, the year, the full year 2020 is a little bit exceptional due to the COVID crisis. And therefore, we can expect a lower level of cash upstream in '21. However, going into '23, you see that the level of cash upstream is going to EUR 5 billion to EUR 6 billion, which represents a cover of 1.1 to 1.3x the dividend, which is a pretty good number, especially when you consider our exposure to Asia and emerging markets, which have an impact of 0.1 billion -- of 0.1x coverage. So excluding this exposure, we would be between 1.1, 1.4, which is absolutely a normal level of dividend coverage for a company like AXA. The second question refers to the level of reserves at XL. As you know, this is an exercise which we are doing at the time of the half year and the full year, not in the middle. But I can tell you, as I said, in the 9 months that we consider the level of reserves at XL -- at AXA XL as adequate.

Thank you, Etienne. Scott, on the top line growth ambitions for AXA XL. Can we hear, Scott?

I can't hear him.

Sorry, we can't -- Scott, we can't hear you. Somehow, we need to see how we can turn your sound on. Not yet.

Should I just answer the question? [Foreign Language]

So I would say, Farooq, let's park your question for the moment until we have solved the technical difficulties, and we'll come back to you once I get the sign that Scott is online and you can hear him. I suggest we pass to the next question. And then once the technical difficulties are resolved, we'll come back to that question. Please?

Our next question is from Mr. Andrew Sinclair from Bank of America.

Three for me, if that's okay. First question was on the XL reserves again. Really, just wanted to understand if there's any implicit reserve build going on within the new XL targets, just to understand that. Secondly was just on the baseline assumption for rates, you're looking at the U.S. 10-year at 1.75% in 2023, German bund at 0%, but you've also talked about lower for longer. Just really wondered if you could give us an idea of what sensitivity of earnings you're now seeing to rates with the new business mix. And if you can give us an idea of what the lower bound of 3% EPS growth would be consistent with our rates. And third and finally was just on -- you've talked about stable debt gearing. I just wanted to know how you're thinking about that. Is it the same metrics as before, 25% to 28%? Or have your debt metrics evolved at all?

Thank you, Andrew, for your 3 questions. All of them are for Etienne. And since it's the second time that the reserve issue is being asked, why don't you also comment in your answer on the cash injection on XL. Second one was on the rate assumption, U.S. dollar versus Eurozone. And third one was on the question around the debt gearing, Etienne.

Andy, so on the XL reserves, there is no implicit reserve build, I would say, in the plan. What I say in the 9 months, and I think I can repeat it here, is that we injected capital of around EUR 1 billion recently in the last 2 weeks to reinforce the capital of XL and to be in a position to benefit from the growth in volumes for next year to offset part of the losses incurred with COVID and, most importantly, to be able to distribute dividends in '22 based on the earnings which will have been achieved in '21. And so we are -- we have -- what we don't intend to do is to recapitalize again the company next year. So it's very clear on that. On the interest rate, you are right to say that we have taken a slight increase in interest rate in our assumption. Why? Because we consider that we are coming back to the sort of pre-COVID level in terms of interest rates, which is absolutely natural. This being said, the impact on our projections is minimal. Why? Because we are matched in terms of asset and liability management. And therefore, the impact of the reinvestment in assets is pretty limited, given the long duration of the existing assets. And therefore, if these interest rates were not going up the way we have projected them, I would not say that the impact would be meaningful or impactful on the 3 -- on the 5% UEPS indicated. This might have a small impact on the solvency. But given the high level of solvency creation over each year, we would see that as nonsignificant and not changing the picture and our ambition level. Lastly, we refer to a stable gearing ratio within our range, but without changing from 25% to 28%. It's most trying to have a sort of stable ratio. Why do we project to raise debt? Because our shareholders' equity is increasing in a very natural way because we don't distribute 100% of our profit. We distribute 60% to 65% of it. So our shareholders' equity is increasing. And with that, we can increase the level of the debt. I remind you that if you compare us to the other players in the sector, our debt ratio is certainly one of the lowest level. So we are absolutely comfortable with this level of gearing ratio. And this cash raised is purely and simply a flexibility we have during the plan period.

Sorry, can I just follow up just on the -- on XL reserves, just one final point. You mentioned that you do these reviews at half year and full year. Will you be able to give us any increased disclosure at full year results, just providing comfort to XL reserves?

There will be an update in full year and exactly in the same way we did it for the last full years and the last half year. Absolutely. This is a normal exercise we do every 6 months.

Thank you, Etienne. Let's give technology across the Atlantic Ocean another chance. Scott, Farooq's question on the top line ambitions of AXA XL. Let's give it a try. There we go.

Right. Hopefully it's working. Can you hear me now?

Perfect.

Yes? Perfect. Thank you. So I apologize for our technology difficulties. Yes, on the growth, on the upside or the tailwind on the growth, obviously, with the price increases in the marketplace, we're seeing more and more opportunities that where we think we can generate an adequate underwriting profit. Sort of the headwind on that is obviously the economic impact of COVID-19. So when we did our planning, we thought through, we're thinking low single-digit growth. Now it could be more depending on what happens in the economy. So thank you.

Thank you very much, Scott. Let's move on to the next question.

Next question is from Mr. Nick Holmes from Societe Generale.

Two questions, if you can hear me. First, would you say that you've changed your focus a bit more on to share buyback? I think in the past, AXA has been a bit reluctant to make share buybacks. So would you say that's starting to change? And then secondly, an operating question. With health insurance, where do you see the main opportunities? I'm thinking sort of which countries and what makes AXA well positioned to succeed?

Thank you, Nick, for your 2 questions. So on the first one on share buyback, it is not a change of our approach relative to the past, but we are in a different era now of AXA. So when you look back over the last 5 years, we have simplified AXA. We have focused AXA. And AXA today is 3 things, if you want to make it simpler: one, a very strong positioning in Europe, as you've seen earlier from Jacques and Antimo, which has been the core of the Ambition 2020 and will be the core going forward; secondly, the largest platform of commercial land risk globally with AXA XL; and then we have some very interesting and promising growth avenues in Asia and in for AXA investment managers on the alternative side, in particular. So we are focused. And now it's very much around how can we organically develop this platform, how can we invest in making it even more prosperous. And that clearly means that every additional acquisition that we would consider has to be clearly benchmarked against our own stock. And therefore, we want to have a very clear and very solid financial discipline around that. And obviously, we also mentioned this morning that we want to continue to focus AXA on less geographies, which obviously means the dilution of our earnings. And it is very natural that in order to compensate this dilution of the earnings, we envisage share buybacks in order to iron out this dilution. So it is not a fundamental shift, but it is clearly a new focus in a new era where those instruments do play a more important role. On the second point around health, what is AXA in health today? AXA outside of the U.S. is one of the largest health player globally. We have around EUR 13 billion of premium. We are very strong around the world in 6 markets. And behind those 6 markets, we have more markets that we want to develop. Health is obviously a line of business that after COVID, nobody doubts, has a lot of importance. And the customer desire to be healthier, to understand their health better and also to do more prevention has never been that high. And therefore, I'm very glad that we have spent the last 5 years in developing a lot on the innovation side around health services. It was mentioned earlier, the emblematic example of telemedicine. We had over 500,000 consultations of telemedicine. So every geography, be it Jacques, be it Antimo, be it Gordon Watson in Asia, everybody is focused on developing health. Health is taking a large place both on the insurance side and the service side in the investment budgets. And certainly, this morning, you have seen Antimo talking about the data health platform in Europe, which for us is a very important step to continue to develop a successful model that has already worked well in Germany across the European markets to really make the breakthrough also on the health side of services. Thank you, Nick, for your question.

Next question is from Mr. Michael Huttner from Berenberg.

I admire you because presenting in the middle of COVID and with the background of difficult earnings is not easy. They asked the question on solvency. I had one on cash and then one -- I had one on disposal and the regulators. So on solvency, 18 to 22, that's the one target which really is clearly above what you had before, 15 to 20. And yet your solvency, you're effectively saying your solvency will stay flat. So, a, I'm wondering where does the extra solvency generation comes from because it's not reflected in the IFRS. They're more or less, you say, the same target. But -- and also what you're doing within AXA solvency because it's clearly not going out as the dividend path is the same. So it's EUR 3 billion a year. I'm willing to share it with you, if you like. The second is on the regulators. So yesterday, we heard from -- and management said that nothing will change in 2021. So no accelerator from you, but still Dutch regulator is saying everything on a case by case. The Italian regulator is saying that as well. How likely is it that the French regulator will allow you to do your capital management plans, which clearly I'm assuming you want to pay your 2020 dividend, but it's not clear that the regulator will really allow it? And the third one is on cash. So on cash, you said 2021 will be lower. You're clearly most on the numbers, but I'm really grateful you gave it. And if you could at least say if it's a 2 or 3 in the big figure. And also, if Switzerland, is there still upstreaming from Switzerland? And also, what happened to the difference between cash upstreaming and holding costs? Because if you just do the cash upstreaming of EUR 14 billion, you deduct EUR 11 billion dividend, you get EUR 3 billion extra cash in the holding, it kind of assumes that the holding cost is 0, which would be very sad because you have to pay yourself something. Anyway, those are my questions. And really well done. It's not easy doing what you're doing.

Thank you very much, Michael. We take this feedback very happily. So you asked 3 questions: 2 were on cash, 1 was the question of what happens to the extra solvency generation. Your second question on cash was why is 2021 lower? And what else is there, for example, the upstreaming on Switzerland. And then there was a third question around the regulator around dividend. I will start with your second question around dividend, and Etienne will answer question #1 and 3 on solvency and cash. So as you can imagine, 2020 was a difficult year from a dividend perspective because I am very dissatisfied with the way it has played out that we are supposedly in a harmonized Europe, yet we have seen a very different treatment of the insurance players across the different countries. It has unfortunately led to the fact that we could only pay 50% of the initially previewed dividends, which I really regret because when you look at our solvency numbers, there's no reason to believe that we wouldn't have been able to pay the full amount. It is clear, therefore, that we are doing everything in our capacity to go back to a normality when it comes to next year. And we have started, obviously, that dialogue with the government in France, with the regulatory authority. And when you also look at our accruals, for example, in Solvency II, you see a very clear sign of wanting to go back to a normality. It is true that the regulators are currently discussing their approach for the dividend of 2020 paid in 2021. And you might have certainly also seen, Michael, that the ACPR and the Banque de France last Friday did a press conference -- sorry, the Banque de France and the ACPR, so the French regulator did a press conference last Friday, and the question of dividend was also touched. And I was very happy to hear that they used a wording, which was around how can we go back to a prudent resumption of dividends. So I strongly hope that we get back to a normality. And I will do everything in my capacity, together with my team, to work for this, to fight for this and to get back to normality. Etienne, question on solvency and cash.

Thank you, Thomas. Michael, you're right to say that the solvency generation -- annual solvency generation will increase versus the last previous plan from 15 to 20, to 18 to 22. And this increase is related to the business mix shift we are experiencing from financial risk to technical risk and especially with the acquisition of XL and the disposal of Equitable Holdings. So we are also saying that we target a solvency level of 190%, but we don't say that this will be the solvency level we will have at the end of '23. And actually, we will have certainly a higher solvency ratio than 190%, if you take the assumptions we gave you. And this gives us the full flexibility to do whatever we want in terms of the topics addressed by Thomas a little bit earlier, notably flexibility in terms of share buyback or in terms of financing the growth of our businesses and notably in health and protection. So I think it's about flexibility. The second topic, which is related to the cash upstream, we will not give you the number, Michael, as you know, for next year, but I don't think you were hoping for that. Yes, there will be some further cash upstream from the GL 2020 program in Switzerland again next year. This would be the end of the program of around EUR 0.4 billion. And lastly, the holding customer amounts to around EUR 1 billion a year. They are reflected in the roll forward you are seeing on the page. So we have minus EUR 3 billion holding cost in the roll forward, which are well presented. But if it's not clear to you, please feel free to ask again.

Thank you, Etienne. Thank you, Michael. Let's pass on to the next question.

Next question is from Mr. Andrew Crean from Autonomous Research.

Firstly, could you just give us a detail to the macro assumptions in your base case? You said that interest rates were not a big swing factor. So what are the swing factors on the macro assumptions in your base earnings case? Secondly, the question about pro-cyclicality. If you will test all M&A-based buybacks, as your share prices rise, the effectiveness of buybacks reduces, and that will tend to mean that you do M&A at the top of the market rather than the bottom of the market, which may not be in the interest of shareholders. Could you comment on that? And then thirdly, I just wanted to go back to the question which was just asked. Are you saying that on your normal assumptions, you will be above 190% solvency by 2023 and that you would use that excess for buybacks as well?

Thank you, Andrew, for your 3 questions. I suggest that Etienne is answering the first and the third question, and I'll take the second one. So the first question was about, what are the swing factors in your base case assumptions? Second question was around the pro-cyclicality of M&A. And third one was around the question of the -- what happens above 190%? Let me start with your second question. As you have seen and as I pointed out earlier to Nick, we are now in a position that we have a very focused AXA, Europe, XL and the growth avenues, Asia and AXA Investment Managers on the alternative side, in particular, this enables us to be in a position of being globally at scale. So there is no piece missing. And our aim is very much to focus on organically developing this platform. And therefore, you've also seen relative differently to previous plans that we do not have a specific lever on M&A in our underlying earnings per share projection going forward. This means that we will definitely focus and favor organic developments. But as we said earlier, M&A should never be excluded. And therefore, we set ourselves a very clear guideline for financial discipline. M&A needs to, a, make sense in terms of increasing those 4 pieces I told you beforehand, Europe, XL, Asia, AXA Investment Managers; and secondly, it needs to be accretive when it comes to the EPS and needs to also present sufficient synergies. And therefore, I can assure you one thing today that we mean this topic around financial discipline and strict financial discipline in a very important way and that we focus rather on organic development. But obviously, you can never exclude that opportunity come your way. But if they come your way, they are very, very strictly benchmarked against our own share price. And what is also important is to see that when you look at the Solvency II definition, the 190%, 50 points above the base of 140%, is not linked anymore to share buybacks. So we are going away from the range that we previously had of 170% to 220%, to a target rate of 190% and no links to share buyback anymore. So we have more flexibility, and I would see it in this context. Etienne, if you could please talk about the number one question, swing factors; and number three, the assumption above 90% on share buybacks.

And I will start with the question number three because there is a link with the flexibility you were referring to. And if you look at the solvency, the sort of normalized solvency creation from 80 to 22 points, if you deduct the dividend, which is around 12, you see that we increased by 6 to 10 points solvency each year. And this is very much aligned with the flexibility we have as well in the cash, as I said earlier. So we have flexibility for share buyback and to finance the growth. But this is the reason why we were saying already in the 9 months that we see that there is a solid case for us to demonstrate that we can pay a dividend not only in '23, but also next year because we think that from a technical point of view, when you look at all the aggregates, profitability, solvency, liquidity, we are absolutely fine. And yes, we would end up, if we were not using this flexibility, with a higher solvency ratio than the 190%, which is the target. And maybe a last word also on this target. Again, 190% after having lost 20 percentage points of market impacts due to the low interest rates and higher spread level is a very, very solid and high level for solvency in that context. Coming back to your question number one, which is what are the swing factors, if not interest rates. Actually, the swing factors are more related to the economic and macroeconomic environment as a whole. So we looked at the GDP, we looked at the projections of the GDP for each of the countries where we are. And we say that depending upon the momentum, the economic recovery, we would have very different top line momentum in our different business lines. And this is certainly a factor #1 versus the projection of the interest rates. You know that in our UEPS projection of 5%, 2% comes from the top line. So this is important. But the top line also has an importance on the other factors, notably on the cost effect. If you have lower volumes, of course, you cover less your cost. So these are really the swing factors, the GDP growth. And in our central case, for the Eurozone, we anticipate a GDP level in '23 equal to the '19 level. It would be maybe in the beginning of '23, but there are also adverse cases where we could say, well, in '23, we'll not be at that level still. Future will tell. We are pretty confident at the moment that our base case is certainly the best we can do at the moment, therefore, our confidence in the plan.

Thank you, Etienne. Thanks, Andrew, for your questions. Let's go to the next question. We don't hear anything. Is there a next question on the telephone?

Written questions now.

Okay. Exactly. Let's move to the written questions, and then we go back to the telephone questions. Please read out the first written question.

The first question is from William Hawkins from KBW. Could you give more information about the ability of your agents in France to operate remotely digitally? Can they all now sell all products in a fully digital structure? If not, what percentage of your agents can sell fully digitally?

Thank you. And thank you, William, for your question. Jacques, that's one for you.

Okay. Thank you very much for the question, William. In France, we have put in place a large program since the beginning of the 2010 in order to digitalize the way our agents work. We have done it through all the interaction through with the clients. We have now an NGM tool, which permits to -- for the clients to make claim, to pay premium, to have a digital interaction with their agents. And all that is well spread in all our agents now in France. Concerning the product, we have also the capability to buy products through Internet for all our retail products, and that is a part of our business. This business was 10% in 2015. It's now roughly close to 20%, so what we call the digital initiated business through Internet. Now our next stage, our next step will be to digitalize the commercial line activity. And because we consider that we -- due to our size, due to our expertise, we are able to create the first digital tool for commercial line in France in order to be able to underwrite more business in better technical conditions. And that has start 4 years ago with a program called [ Oz ]. I have this morning detailed this program or give some flavor about this program. We are right in this middle of this program. It means that it concern only 55% of our new commercial line business in France, but we intend to grow this program to new products. We have already 8 products on this program today, digital. We will add in the 2 coming years 8 more products, and that will concern in 2023, 75% of our new business commercial line. We will have to train agents to use this tool. Up to now, we have roughly 1,000 tied agents that are -- that on a daily basis use this tool. We will have by 2023 around twice much, 2,000.

Thank you, Jacques. Let's try and go to the next written question.

The next written questions is from [ Maximilian Vaults ] from [ VW Hoiter ]. There are 2 questions, actually. What are -- the next question, therefore, is, what are your plans in China? How will you compete with Ping An and China Life Insurance? And the next question is, what are your plans for Germany? How will you increase the results?

Thank you very much, Maximilian, for your 2 questions. I suggest that Gordon Watson will take the first one on your -- on our plans to compete well in China. And Antimo Perretta will take the third one on -- sorry, the second one on Germany. Gordon? I think we can't -- ah, there we go. Go ahead, Gordon.

Thanks, Maximilian, for the question. The great thing about China is it's large. And as such, there's a massive protection gap, and it's not a zero-sum game. And we have terrific assets in China. We have AXA ICBC, we have AXA Tianping and we also have AXA Hong Kong when we look at the Greater China strategy. So I'd just like to give you some context in terms of where we are in China and where we are in Asia at the moment. If we look at Asia first, we are up 3% at 9 months in GWP. And if you look at health, which is where we really want to compete, we're up 10%. And if you look at our underlying earnings, we're up 7% in the half year. And if you look at health specifically, we're 13%. And of course, to do what we want to do and really grow our business, we need distribution. And if we look at our distribution, we can see the number of agents is up by 14% over the prior year, which is quite remarkable in this kind of year to attract agents and have them both digital sales, digital sales. And China is becoming more normalized, and they can do this. And the other thing is the Net Promoter Score. If we look at all our markets in Asia, our Net Promoter Score in every market is either at the market or above market. And this is very important in having the right people. And if I go to China specifically, China is starting to come out of and be a lot more normalized than many parts in the world. And we have a very solid China strategy. So when we look at ICBC, which is our life license, and we have a joint venture and they're the biggest bank in the world, our half year underlying earnings there was up 13%. And the number of agents, and we're really adding agents to supplement our bancassurance distribution, is up 19%. And what I'm really happy about is the quality of the business. The regular premium is up 16%, which shows that we really truly are creating value in China. In terms of our key strategies in China, specifically, I just want to say, first of all, we start in the platforms. And we have [ Swede ] and we have [ Zurich ]. Secondly, we work with the government, and we just -- it was a massive case 2 months ago, where we now have a million customers, and basically nobody else in the world can do that. So we are using our global expertise, our expertise from Hong Kong to really differentiate. But I have to say, end point is it's not zero-sum game in China. I'm very happy with the huge progress that we made in the country. So thanks for the question, Maximilian. I hope that answers it.

Thank you, Gordon. Let's move to Maximilian's second question on Germany, Antimo.

Yes, thank you for the question. So in Germany, we will do that in 3 layers. I would say, first, we will continue to grow in our preferred segments. We have done this in the last plan, and we'll continue to do in the new plan, in the health business, growing much more than 5%; in commercial on more than 2%. And we will continue to stay really and grow also, but to be profitable in the retail business. We're doing that with our tied agents. We have really a lot of innovation in the way that interact with the customers with new digital tool, [ AXA Price ]. And all these elements will help us in a digital way to interact with the customer and to continue to grow. The second element is about we'll profit for the experience in Europe coming from Switzerland, coming for the other areas where we have done a lot in our in-force book in the life. We will continue to work on that to make sure that we operated in the right way in this Life & Savings book and make sure that we increase our profitability. And the last element is about costs. We have an ambition plan to cost savings. We'll continue to do that with a lot of really deep elements that we will put in place. And I have seen this in the last plan, and we'll continue to do that and be more simple to redesign our customer journey in reducing complexity and go much more to the cloud. So all these elements will help us to increase our position in Germany and to increase our earnings in the couple of next plan in the couple of years.

Thank you, Antimo. I suggest we move back to the phone line. And as I assume that we did not all hear the answers to Andrew Crean's 3 questions, I would just like to quickly come back on them again. One was the question around what are the swing factors in our case? It is clear that the swing factors are very much related to GDP development and the macro environment. Your second question, Andrew, was about the pro-cyclicality of M&A. My answer was very clear. AXA is now focused. AXA is globally scaled. We want to put the majority of our efforts in organically developing that platform, but we will put ourselves under a very strict discipline of comparing M&A opportunity against our own return of the AXA share price. And then thirdly, it was a question around what happens above the 190% of solvency. Etienne said that share buybacks are not linked anymore to 190%. And with the projections that we've given you and what is still to be coming, so for example, the integration of the internal model of -- sorry, the integration of XL into our internal solvency model will clearly bring us to 190% in a period in which interest rates are even lower. So the 190% today mean much more than 190% 2 years ago. Let's go on to the next question on the phone.

Next question is from Mr. Peter Eliot from Kepler Cheuvreux.

I'm guessing you didn't hear my questions first time, so I'll repeat them quickly. But the first one was a follow-up on Michael's question on the regulatory outlook because I guess some of your peers have been prevented from paying dividends in full this year. But many have said we'll pay like a double dividend or 1.5x dividend next year. And I'm just -- obviously, you haven't said that for this year. But I'm just wondering if the regulator were to say you couldn't pay again next year, might you consider sort of escrow account or a commitment that shareholders would get it back as soon as you were allowed to? That was the first one. The second one was a follow-up on Andrew's question on the UEPS. Because I guess M&A is not the same focus, but hopefully, the money that would have gone towards M&A is still available for increasing UEPS in one form or another, whether it's through share buybacks or another means. So I'm just wondering, in that context, whether there should still be a bucket there for M&A or equivalent? And then finally, I wanted to ask on XL. Just wanted to understand your view of the profitability and volatility a bit more because if I look at Slide 23, then I'm just struggling to reconcile 30% of the business being at 98% and 70% basically being under 96% with a group -- with a target of 96% overall for the business. It seems you should be able to achieve less than that, but there must be some businesses achieving very good profitability. And to help us sort of think about the volatility, I'm just wondering how we should think about the volatility now compared with the charts you've shown before of the sort of 1 in 10, 1 in 20 year events. Just wondering whether that's come down further.

Okay. Thank you very much, Peter, for your 3 questions. I suggest that I will take the first 2 on the regulatory outlook and the question on M&A and that Scott will answer the third question around Page 23 and which line of business has really moved in which area. And if necessary, I would also like, after Scott has spoken, that Alban could maybe quickly add something around our risk appetite because these numbers are obviously also very much linked to, has risk appetite changed, which it has not. And so let's do it this way. So on the first question, regulatory outlook. Look, our focus is very clear, and I want to repeat again what I said earlier, we are pushing and we want to get back to a normality in 2021 for the payment of 2020. We have, unfortunately, had to live through a situation in which there was very fragmented and different treatment of dividend payments across Europe despite the fact that the solvency numbers, if I speak for AXA, would, under any circumstances, allowed a full dividend payment. So our focus is very much getting back to a normality. You've seen in the accrual of Solvency II the dividend we are thinking about when it is around the accrual. And all we want to do is talk to the various stakeholders, be it the government in France, be it our regulator, which we are already in the process of doing to get back to this normality, and the Board will decide on this by the end of February. And hopefully, my aim and our aim to get back to this normality will become true. And as I mentioned earlier, I was very positive during the press conference of the ACPR and the Banque de France that there was a clear opening around this for 2021. On the second piece, what is the focus of M&A? What is it not? Again, we have a scalable and scaled platform at AXA across different geographies. Our focus is very much to invest and build this platform organically. And you've seen in plenty of the presentations this morning, concrete projects to do it. We obviously will continue to look at M&A, but this is not our daily focus. Our daily focus is really to make sure that we allocate the necessary money for the organic development. And therefore, we decided not to have a separate bucket on M&A and also not to put ourselves under pressure and not lose focus. On the third question, Scott, if you could talk about Page 23 again and the different components of the portfolio when it comes to combined ratio. And I would like Alban to comment afterwards quickly on the question around risk appetite, which has not changed. Scott? We can't hear you, Scott. So I think we've got a technology issue again across the Atlantic Ocean. Maybe, Alban, you start with the risk appetite. And then hopefully, by then, we have sorted out Scott that he can talk about the Page 23. Alban?

Hello, everyone. So we have not changed our risk appetite on net cat, which, as you know, is a EUR 500 million deviation for 1 in 20 years, net of reinsurance and net of tax at group level, not at tax level, at group level. That being said, as you saw in the presentation, we now assume more prudent cat load at AXA XL, moving from 4 to 6 points of combined ratio, which means basically that we now have a higher certainty to be in our risk appetite as was stated before.

Thank you, Alban. And let's try again with Scott. These are obviously the difficulties when you try and do an Investor Day in COVID, but we won't stop trying. Scott? No, it doesn't -- still doesn't work. So I suggest, Scott, we can't -- now we can hear you.

Well, am I there? Can you hear me now?

Yes.

Oh, good. All right. Yes. So yes, again, I apologize for the delays on the technology. Yes, it's an interesting question with our chart in terms of where the products are in terms of profitability. And we look at profitability and we look at volatility as 2 slightly different things while related. And we're actually very proud of the work we've been able to do in the last year in terms of having 70% of our portfolio at or below 96%. And what that -- sure, within that portfolio, there are numbers a lot closer to 96% and ones that are doing a lot better. And we manage that business. And the first step and say, okay, look, let's get all of our products profitable. At the same time, in terms of managing volatility, we also want to make sure that from a line-sizing standpoint, that we're not exposing ourselves to our lines being too big. So we did both of those activities. But -- and they're related, but it's really important that we get these products profitable. And as Alban mentioned, we took a little more conservative view of our cat load this year just to help out a little bit on that volatility. So thank you, Thomas.

Thank you, Scott. And we'll move on to the next question.

Next question is from Mr. Jon Hocking from Morgan Stanley.

I've got 3 questions, please. It was interesting to see that the Solvency II capital generation target has gone up. I wonder whether you could comment a little bit in terms of the cash conversion of IFRS earnings and whether that should go up commensurately. And linked to that, what payout ratio are you penciling in for XL in 2022? Should we assume that, that is sort of broadly in line with the group payout ratio? Second question is, why have you only given 1 year earnings guidance for XL, given the sort of portfolio remediation that's happened and the pricing? It seems like the visibility there should be a little bit better. And then finally, just in terms of the higher cat loading guidance for XL. I think, Scott, in your comments, you mentioned, so empirically looking back at what you've seen in recent years, you had a higher cat loading. Have you only put in place additional portfolio actions, I think, in terms of line sizing and higher reinsurance protection at the beginning of the year? And to all extents of risk, are you actually just being very, very conservative in terms of the extra 2 points? How actually we could come back and revisit this once you see the portfolio management actions work through?

Thank you, Jon, for your 3 questions. I suggest that the first 2 are being answered by Etienne and the third one by Scott. The first 2 was around the question of the Solvency II targets, cash conversion of IFRS earnings and then the payout ratio for XL in 2022. Second question was why only a 1-year guidance on the XL earnings, both of these questions for Etienne. Third question for Scott around the portfolio actions, will you revisit those in terms of line sizing and so on, once you have seen that the portfolio is in a space where we wanted to be. Etienne, for the first 2?

Yes. Thank you, Jon, for your questions. First of all, about the conversion of underlying earnings into cash, usually, we have a ratio of 80% to 90% of P&C business. We have 60% to 70% on the Life business. You might wonder why such a lower level. It's basically composed of 2 things: first, 70% to 80% on the major countries; and rather 50% on Asia and emerging markets. So this was for the first question. The second question is the one -- only 1 year of earnings for XL. First of all, this target is already existing. This was the target for this year. And therefore, it's a natural target for '21 because the COVID disturbed a little bit the performance this year. But I think most importantly, the target for next year for XL is not the ultimate target. Why? Because the pricing effect will continue to play over in 2022. So you might expect, everything being equal, a further upgrade in profitability beyond '21 for XL.

Thank you, Etienne. Let's move to Scott on this note around portfolio action. Will you revisit them, yes or no, once we are in the green zone, i.e., above EUR 1.2 billion? Scott, we don't hear you again. We don't hear Scott.

Actually, while we're waiting for Scott, can I just come back on Etienne's comment about the capital...

Yes, exactly. Jon, use the time until we sort out the technology. Jon, what was your question on Etienne coming back?

Yes. Just to go back up to the -- you mentioned there was going to be a dividend from AXA XL in 2022 based on the '21 earnings. Can you comment whether that's likely to be in line with the group payout ratio? Or is it likely to be a sort of a conservative start to cash coming out of that unit again? Is it likely to be the sort of 55% to 65%? Or is it going to be a different level?

Etienne?

No. Look, I will not give a comment on a given year. But on a normalized and ongoing way and if you look at the perspective, '23, there is no reason to see XL not distributing an equivalent level of dividend compared to the other P&C businesses of the group.

Okay. So over the course of the plan, we should get to the 80% to 90% payout ratio for AXA XL?

Exactly.

Scott, shall we try again on Jon's third question?

Sure. Okay.

It works.

Sure. Can you hear me?

Yes. Go ahead.

Okay. In terms -- wow, it's getting a lot of echo. There. How about that? Can you hear me, Thomas?

Yes.

Okay. Thank you. In terms of volatility, we look at our line sizing, and we put EUR 25 million to sort of our maximum net line. Obviously, we'll have products below that. And we look at them on a regular basis to determine the appropriate retention based on product profitability, opportunity, et cetera. But we're not expecting dramatic changes in the net lines. In terms of the cat load, the 4% to 6%, we put that in as we looked into more recent cat years, what's been happening is there's been a series of events, they've been large, but they haven't been significant. And so we felt that there's been a lot more activity around sort of this almost attritional cat business that adds up over time. And we felt that would be better to put a little bit higher cat loss estimate into our plan because that seems to be what's been happening the last few years. We always will evaluate that every year and take a look at it as to what's going on, but we felt a little more prudency on the cat load made more sense for 2021.

Thank you, Scott. Excellent. Let's move to the next question on the phone. I can't hear them now. Next question on the phone?

The next question on the phone comes from James Shuck from Citigroup.

I'll try and get these out quickly before we have another technological failure. Just on the ROE, you have an ROE target of -- it used to be 14% to 16%, now down to 13% to 15% over the next 3-year period. You've obviously sold EQH in that period, which would have been a low ROE business. So just keen to understand some of the moving parts around that load target, please? Secondly, wanting to return to the cash conversion question. So your cap gen on 18 to 22 points would be about EUR 7 billion to EUR 8 billion, and you're guiding towards EUR 5 billion to EUR 6 billion of free cash flow in 2023. So the cash conversion on the cap gen is around 70% to 80%, which looks low. I appreciate that Asia would be a low conversion, but there is an element of funding growth already in that cap gen number. So just trying to understand why that conversion is low. And then third question, I just want to be crystal clear about the second tranche of the 2019 dividend, please, because in the EUR 11 billion dividend roll forward in the slide, that doesn't seem to imply that the second tranche gets paid. Why would that not be the case? I appreciate the ACPR has put you under some pressure and suspended things. But there's no reason why, if your capital insolvency is as strong as you say it is, that you shouldn't return and pay back that second tranche when you are able to do so.

Thank you, James, for your 3 questions. I suggest that Etienne is going to answer the first one. So the first one was on the return target 14% to 16% down, which is obviously due to a change in methodology. Second question was on the cash conversion, if you can give more detail around the 70% and 80%. And I will take the last one on the second tranche of the 2019 dividend. Etienne?

Yes. So the ROE, we have changed the definition. Instead of taking an adjusted return on equity, we take an underlying return on equity. Why? Because we think we have too many KPIs, too many metrics, non-GAAP measures, and would like to simplify our communication on the underlying earnings on the one side and net income on the other side. And therefore, because we don't benefit from the contribution of the net capital gains, which are between EUR 300 million and EUR 400 million a year, mechanically, the return -- underlying return on equity is 1% lower than the adjusted return on equity, nothing else. So no change, it's really equivalent. And in the same way, you saw that we base our payout ratio on underlying earnings and not anymore on adjusted earnings, exactly for the same reason. It's to simplify our communication, reduce the number of non-GAAP measures. So this is absolutely consistent. On the cash conversion, I just reiterate what I said is that there will be a lower level of conversion next year because of the very exceptional year we are going through in 2020, and then we'll go up by '23. And I will stop there. I think when we say we're going to give an indication of EUR 5 billion to EUR 6 billion for '23, obviously, our best estimate is EUR 5.5 billion. Our aspiration is EUR 6 billion. If we had an aspiration of EUR 6 billion, then we will have a dividend coverage ratio, as I said earlier, which would be at 1.3x, which looks to us like a very reasonable number.

Thank you, Etienne. And then, James, on your third question around the second tranche of the 2019 dividend. As I said earlier, our focus is very much to get back to a normality of dividend payment by 2021 for the year 2020. This is our absolute focus in the discussion with the regulator and also with our Board. The next important milestone in this is the Board discussion and decision in February of 2021, once we also know the full year numbers for 2020. But I would like to draw your attention again to the accrual of the dividend that we have today in our solvency numbers, which is a dividend level of EUR 1.43, which is clearly the dividend that is beyond the range of the 50% to 60% payout that we normally would have applied. And therefore, the focus is very much on the February Board on the decision of getting back to a normality, and this is the first milestone, which I think is a very important milestone. And what happens afterwards, we will see, but we should not tackle too many issues at the same time. Let's move to the next question.

Next question is from Mr. Michael Huttner from Berenberg.

Fantastic. These are really little questions. First one is on AXA Life Europe. If you can give us an update what's happening there. The second is on the implicit kind of group combined ratio, so 93% for the group in 2023. XL doesn't sound as different. It sounds to me as XL is going to use excess cash to rebuild the reserves from an adequate to the more like AXA level. So I'm assuming that France and Europe will be somewhere between 90% and 91%. And I wonder if you can maybe comment on that. And then the last one is on the starting point for all these numbers, so 2020. If you could give us a figure that would be best. But if you could actually give us figures to allow us to work it, that would be even -- that would be second best but still lovely, which is basically, what is the figure we have to exclude to reflect COVID? And what is the figure we have to exclude to reflect excess natural catastrophes?

Thank you very much, Michael, for your second batch of questions. When you say they're little questions, they will be big questions. And you have proven that they will be big questions. So on AXA Life Europe, Etienne is going to give an update. On the group combined ratio, Etienne, you will also talk about XL versus France and Europe. Because as Jacques and Antimo have shown in their presentation earlier, France and Europe is also down by 1 point. And then thirdly, the starting point, 2020, I have to remind you that we are not today in the presentation of the full year results of 2020. We are in the Investor Day that is well ahead of this presentation. So when you think about excluding COVID, you have already some very clear numbers, and Etienne is going to talk about them again. So we can give you some outlines, but please bear in mind that the full year 2020 presentation is happening in February and not on the 1st of December of 2020. Etienne?

Yes. Thank you, Thomas. And thank you, Michael, again. So maybe starting with the third question to make the link with the comments of Thomas. Obviously, you know that we communicated on EUR 1.5 billion impact net of tax and net of reinsurance for the COVID in 2020. So this would be one of the numbers to be used to come back to the normalized level, the other one being mainly the nat cats, and we'll see at the end of the year to which number we are coming. Coming to your question about AXA Life Europe. As we said in the half year, we are looking at alternative solutions, which are mostly reinsurance solution. So it's -- what we want to do is very concrete, but it's not -- it's still early stage because we go through a certain number of regulatory approvals, and we'll communicate in due time when it will be over. But we are reasonably confident that there will be some positive news on that front. Lastly, on the combined ratio, 93%. What I can tell you is that across the board, all the P&C operations should improve their expense ratio by 1% as a result of the cost-efficiency program we have introduced to you this morning. And lastly, XL will, of course, have a contribution to this 93%. But the other way to look at it is on a normalized level -- purely normalized level, this would be a sort of 2% improvement versus the current year level, with minus 1% coming from the improvement in the loss ratio in commercial lines, mainly contributed by XL, and the second 1% coming from the expense ratio.

So I suggest we'll move again to the webcast because there are still quite a few questions, and we should try and balance the phone questions and the webcast.

There are 2 written questions from Colm Kelly, UBS. What is the rationale behind removing the operating free cash flow target that was 1 of the 4 core targets within the previous Ambition 2020 plan? And can you articulate the management actions that are guided to support the EUR 14 billion cash upstream target?

Thank you, Colm, for this question. So Etienne, on the first one, the why replace the operating free cash flow against cash upstream. Obviously, it's much simpler. And secondly, the question around the management actions of the EUR 14 billion. Go ahead.

So you already answered, I think, Thomas. It's, again, simplification. We were before giving an operating free cash flow target, it's absolutely true, but it was not sufficient. And therefore, we had to give also a percentage of conversion into cash of this metric. Therefore, we had to give 2 metrics. At the end of the day, you obtain with that the cash upstream. So let's focus on the end result, which is the cash upstream. This is what matters for our shareholders because you want to look at how much cash is coming to the holding and what is the percentage of this cash, which covers the dividend. That's it. So very simple. On the in-force measures, we have 2 categories of in-force measures. We have the ongoing in-force measures, operationally driven, which are looking at how can we help our customers convert into other products. But for this, you need very attractive new products, alternative proposals, which make sense because you need to advise your customers in their best interest. So this is one thing, which is being performed in a lot of different countries and which have certainly been accelerated since the last plan. The second stream of actions are more portfolio actions driven, like what we did, there was a major buyout of policies in Switzerland -- sorry, in Belgium, called -- on the product called Crest. There was a second measure, which shows the GL 2020, the famous tale of 2020, which was a conversion -- a change in model in the Group Life in Switzerland. And lastly, there was the sale of Group Life portfolio in Germany called Pro bAV. So you see that we have been touching various countries, and we will continue with the same state of mind to work on these measures. And the opportunity as well to remind that cash upstream, as was said by and presented by Thomas, will be -- the weighting of the cash upstream in the bonus of the managers of this company will be increased.

Thank you, Etienne. You said there was a second question on the webcast. Let's focus on that second one.

Is it -- there's a question from Thomas Fossard, HSBC. There are 3 questions, actually. GA savings spread, what gives you confidence that you can maintain the spread unchanged over the duration of the plan? Then there's a question about COVID-19 claims estimates. Can you confirm that you're still looking for a total cost of EUR 1.5 billion in fiscal year 2020? Third question, restructuring charges. What level of restructuring charges do you expect on a normalized basis per annum over the next 3 years?

Thank you for that question. I guess, Etienne, those 3 are for you. So what makes you confident on the spread on the GA savings to be unchanged? The COVID claims confirmation, which you have already given for 2020, EUR 1.5 billion, you can do it again, and the restructuring charges.

Thomas, thank you. On the GA savings, what we said is that our cash flows are being matched and the level of reinvestment is pretty limited over the next 3 years. And so this is the first thing. The second thing is that the crediting rates are going down as well. So the combination of a very long duration of assets and flexibility on the crediting rates give us this confidence that the level of margin in Life will be hardly hit. COVID, I think, is clear. Restructuring charges, so we want to reduce our cost by EUR 500 million. There will be, of course, restructuring costs in relation to this amount split over the next 3 years.

Thank you, Etienne. Let's move back to the telephone line, and I hear that there are 2 questions still outstanding. One is from Oliver Steel and the second one is from Pierre Chedeville. So Oliver, why don't you start? Oliver Steel?

I'm afraid it's a written question.

Fine. We take it in any shape or form, written or oral. Go ahead.

Three questions actually from Oliver Steel from Deutsche Bank. First, can you provide more detail on reinsurance cover against nat cats? What is the maximum loss that you now expect relative to nat cats of different severity? Second question, what proportion of profits now derive from noncore operations? Most of the businesses you've sold recently don't actually contribute nat profit, which would suggest that the commitment to neutralizing the EPS impact from disposals is not that large. Third question. Should we still be expecting any capital release from AXA Life Europe? Is this included in the EUR 14 billion of cash remittance?

Thank you, Oliver, for your 3 questions. I suggest that Alban will take the first question around the reinsurance coverage on natural catastrophes and the maximum loss. Etienne will talk about the profit of the noncore and what does it mean for the question of neutralizing EPS. And if I have to say, quite a few of the operations that we have sold have actually been very profitable. Therefore, the issue of the dilution is a real issue. And then thirdly, Etienne, if you can also talk about the capital release of AXA Life Europe, and is it included in the EUR 14 billion, yes or no? Alban?

So the question was a bit difficult to hear. So I hope I will be answering it. So as always, cat protections are made of different instruments, such as the excess of those aggregate quota shares and cat bonds, and we use those different instruments and we arbitrate between them in terms of costs to us. The important thing is that for 2021, the market is hardening, and therefore, prices are up. We are able to contain those price increases. And in fact, we think that our cat reinsurance cost is fully flat, thanks to the fact that XL is writing less cat business or cat-exposed business and the fact, as I said, that we are optimizing between those instruments. And overall, as I said earlier, we maintain at group level our risk appetite.

Thank you, Alban. Let's move to Etienne with the 2 remaining question of Oliver's.

Yes. Thank you, Oliver. On AXA Life Europe, I say at the half year that we were considering a few hundred million euros of cash upstream. I cannot tell you more than this at this stage, but it gives you an idea. Second, related to the proportion of noncore, I would take as a proxy -- you know that a lot of the noncore businesses are within the International segment. So you can take a percentage of this as a proxy. We will not give you a precise number. But you can imagine that if you take a very simple rule of thumb, for each EUR 100 million of business -- of earnings contribution from a business, which is being disposed of, you multiply it by at least 10, you have around EUR 1 billion of cash. So I will not tell you how many times this will be -- if it will be 1x this, 2x this, 3x this. But this is what we are talking about.

Thank you, Etienne. Let's move to the next question, which I guess is from Pierre Chedeville.

From Pierre Chedeville, CIC. Three questions, actually. First of all, considering natural inflation of cost, what is the real effort of cost reduction? Second, regarding the EUR 1.2 billion target for AXA XL, could you give an approximative breakdown between insurance and reinsurance? Third, how do you see the growing interest of the banks, notably in France for commercial lines?

Very good. Thank you, Pierre, for your 3 questions. First one, for Etienne, on the natural inflation of cost. Second one, for Scott, what is the repartition insurance-reinsurance of the EUR 1.2 billion? And third one for Jacques, what is the interest of banks in commercial line insurance, and what does it mean for AXA in terms of competition, but also in terms of partnership. Etienne?

So on the cost, the EUR 500 million are net of inflation and net of growth. So we don't make here -- we don't need to give an indication for the inflation. But if you were making a rule of thumb and saying, I might expect a 3% -- let's say, a 3% inflation per annum, this would mean EUR 1.5 billion savings.

Very good. Scott, on the second one, repartition reinsurance-insurance?

Yes. Rule of thumb of reinsurance is approximately 25% of our business. So you can use a 25%-75% as a proxy on the EUR 1.2 billion.

Thank you, Scott. Jacques, on the question of banks in France and commercial insurance.

So very rapidly, commercial lines, as you know, needs expertise. And those who have started to go on this field without any expertise, usually they fail. And if you look at the different entities, up to now, you can -- that attests to what I am saying. So that means what? That means that, first, we have to strengthen our differences and our -- the fact that we are ahead in the market. And that is all what we are doing with the help of our agents about digital tool, training, expertise, thanks also to the XL proximity because we work very well with XL on that stuff. Secondly, concerning bank, I think that the easiest way to go there is to make partnership. And we are -- we have some partnership with them. It -- on the, let's say, technical side, it's worked well. On the commercial line, it's -- on the commercial side, it's -- for banks, it's difficult to go on the market they don't know and where they are not expected by the clients.

Thank you, Jacques. Let's move to the next question on the webcast, which is from [ Tony Leung ]. And I guess it's probably a question around Asia.

Yes. Can you please share further details on your Pan-Asian strategy? Which markets do you foresee contributing the most growth? And when do you project MCV sales will return to Hong Kong?

Thank you, [ Tony ]. That is clearly a question for Gordon on what is our Pan-Asian strategy and when do we expect MCV sales to come back. Gordon?

Thanks, Thomas. Yes. Let me take the last one, first. Clearly, Hong Kong has this prolonged closure of the mainland border, and it's very frustrating because now we're going through the fourth wave. But the Hong Kong business is doing well. The domestic business is strong. It's about flat this year. And if we look at the other business we do, it's employee benefits. In employee benefits, we are the market leader in Hong Kong, and we have 20% market share. I also see positive steps with the GBA when that happens, and we will open a service center there. But on your specific question, we really see a pent-up demand on the MCV business. We know it's there. Our agency force know it's there. So we do think that, that will come back relatively strong. So we don't know when exactly with the fourth wave, but I think we have to be patient, but hopefully sooner than later. And when we talk about our Pan-Asian strategy, I gave you some numbers earlier. Hopefully, you heard them. There was a bit of an echo. But if we look at what we're doing in Asia, we're growing our business strongly. The GWP is up in Health at 10%, and Health & Protection is our focus. And when we look at the countries, we really divide them into 3 areas. One is Greater China, which is ICBC-AXA, Tianping and Hong Kong, which, for me, really has a massive potential to grow. That's the one we're really looking for a lot of growth there. We have Japan. But if you take out the Health business in Japan, although it's a saturated market in many ways, but not the Health business. There is a big health protection gap in Japan. And we've done very well this year. For the last 6 months, we have been the #1 foreign insurer in terms of sales and #5 overall, and that really is a first for AXA in Japan. So we're very robust there. We have some great CEOs: Seiji in Japan; Yamin, AXA Tianping; and Madam [ Wu Xian ] in ICBC-AXA; and Sally Wan in Hong Kong. And then in Southeast Asia, for me, that also is a growth engine. We have amazing assets there. We have joint ventures with 3 of the largest banks in Asia: Mandiri, they have 22 million customers we really want to focus, and the customers we do have, adding health writers; and we want to do that in Thailand and the Philippines too, where we have Krungthai Bank, and we have Metro Bank in the Philippines. So overall, we see growth in all of our markets in Asia. Longer term, clearly, China is going to be the biggest market in the world, and you know our strategy is very clear. It's to look at China. It's to really transform our distribution. It's to pivot towards Health & Protection. It's to grow our brand, which is what we're doing, and it's to have the best people that we possibly can to lead the countries. And I'm very happy with the CEOs we now have in place. They're very diverse, both gender-wise, both country of origin. So we have a great team to really tackle the protection gap and the opportunity that presents itself in Asia. So thanks for the question.

Excellent. Thank you, Gordon. And let's move to the last question, which is a question from Thomas Fossard.

Could you provide any update on the benefit of moving AXA XL to the group's solvency internal model?

Thank you very much, Thomas, for this question. Alban, this would be a great question for you since we are in the final round of the negotiations around the integration of AXA XL into the internal solvency model. Alban?

So effectively, we would have the answer in the next few weeks. And we are still very confident that we will achieve the 5- to 10-point gain that we announced.

Thank you, Alban. Unfortunately, we have to come to an end of this Q&A session. I know there is still quite a few questions left, which encourages us all. And please feel free in the meetings and discussions that we'll have in the next couple of days to address all your questions because we are really here to answer all of them. I want to thank you for being here this morning, and thank you certainly your very good questions. Definitely, thank you also to all my team members to have been here and to answer all the questions. I want to apologize for the technical difficulties that we had in the post of that IR Day. But as you can imagine, in a COVID-19 situation, it's not always easy to get everything perfectly done on the technological side. Nevertheless, I hope you have really understood most of it. And for us, it was very important to have that event with you despite the fact that it was challenging on the technology side. If I summarize again, what is our Ambition 2023, Driving Progress Around? It is clearly oriented around 5 key strategic actions. Number one is we want to really make sure that in a post-COVID environment, we benefit and capitalize on the strong customer demand increase around Health & Protection, which is growing, Health & Protection across all geographies, you heard this from everybody this morning, making sure that we don't only do this on the insurance side, but also on the services side. Our aim is more than 5% growth on health insurance. Second key strategic action is around how do we continue to simplify the customer experience. We have some great examples in the group, AXA Belgium, for example, where we have gone very far in almost digitizing 80% of our processes. We want to continue this journey, making sure that we marry the physical assets that we have, our distributor, our agent, with the digital technology to become one and be there even more for our customers in a more simple, more faster and more transparent way, all of this being accompanied by our ambition to reduce our cost base by EUR 500 million between 2020 and '23. Third one, continue to strengthen our underwriting performance. You've seen our clear aim of getting to an all-year combined ratio of 93%. All geographies will contribute to it. However, the largest jump is clearly being allocated to AXA XL that is operating in a great, important and increasing rate environment. Fourthly, we are the first insurance company, and we realize that it's extremely important to put our societal engagement at the core of our strategy. So the fourth priority is very much how can we sustain, how we continue to remain the leader on climate change on the insurance side and on the investment side. And we have set ourselves a new goal, which is how can we reduce the carbon footprint of our GA assets, general account assets, between now and 2025 by 20%. And then lastly, the fifth priority, and I guess, for you, the most important priority, how can we make sure that we grow our cash flows across the entire group, which is obviously making sure that the cash flows of the operating businesses will increase through an increase of the performance of those businesses, but also to continue, as Etienne's said earlier, working on and continue to work on the in-force business on the general account, making sure that we continue the journey on simplifying the footprint and making sure that we put ourselves under very strict capital management guidelines, always making sure that when we invest money, we do it in the best interest of our shareholders and we benchmark it against an investment into our own share. I really want to thank you again for being here this morning. Thank you for asking so many questions, and I wish you a great rest of your day. Thank you very much.