Yes. Good morning, everybody. Welcome to NN Group's Capital Markets Day, where we're going to talk about future-ready growth. Welcome to people on the webcast, and welcome to people in the room. It's great to see so many familiar faces with us here today. I already saw Dennis walking around, and he had to enjoy the brilliant Dutch weather today. I'm sorry he had to walk outside to come to the office and get a bit of a pouring down on you. I'm Robin van den Broeck. I'm the Head of Investor Relations of NN Group, and I'll be your moderator for today. Let's talk about the program. First of all, David, our CEO, is coming to the stage, and he's going to outline our strategy. NN IPO'd in 2014, and since the IPO, NN has delivered consistent growth.
David will outline new targets for 2028 that will further be a testimony of this growth journey. David will stay on stage because we know you like him so much. He'll talk about our Future Ready initiative. Last year, NN added being a digital and data-driven company to its strategic commitments. David will explain how our Future Ready program embodies that into our strategy. He'll call to the stage Tjerrie, our Chief Analytics Officer. Tjerrie will give his view on artificial intelligence. He'll show you some use cases that are already live within our company today. We're quite proud of that. After that, we'll have a 15-minute coffee break because we all need a little bit of caffeine.
Then Annemiek will come on stage, our Chief Financial Officer, and she'll talk about how continued business diversification will improve our growth profile over time. She will also commit to our capital return policy, which has compounded such attractive returns for shareholders over time. We will continue to do so whilst keeping flexibility. If you look at our strong net capital build and net cash build, there's ample opportunity to expect more further down the road. Today, we're also putting a special spotlight on Europe, which is the strongest growth segment within the organization. I'm very excited to have two local CEOs with us today, Paweł and Kuldeep of Poland and Romania. Paweł will outline how Poland has been able to integrate the MetLife acquisition and do a digital transformation at the same time to become the biggest OCG contributor of Europe in 2024.
Kuldeep will highlight how Romania has tripled its OCG contribution between 2019 and 2024. From that higher base, it will more than double towards 2028. Very impressive growth in those two markets. After this, we'll have a plenary Q&A session with David and Annemiek. After the Q&A session, we'll end the webcast, and the people in the room will break for lunch. After lunch, we'll have four rotating breakout sessions, hopefully in a friendly, more or less formal setting. One will be with Wilbert, our Chief Risk Officer, about balance sheet management. The second breakout session will be with Frank, and he will highlight how important it is to have an excellent distribution capability at hand to drive further growth within Europe. Tjeerd will do the third breakout session on Netherlands Non-life and the bank.
On Netherlands Non-life, Netherlands Non-life is really leading the adoption of the artificial intelligence wave. We think that will create efficiency gains that we can leverage for growth also in that segment. On the bank, we see a stable and strong remittance pattern ahead of us. The last breakout session, but not the least for sure, will be led by Leon and Marius on Netherlands Life and Japan Life. Now, on Netherlands Life, we are confident that we can continue to remit the stable remittance pattern that we see today. We also do not see a cliff edge after 2040. On Japan, yeah, we are working on the business improvement order. We are hopeful that we can exit that soon. After that, we also intend to recapture market share in that market. Marius will share his views around that.
Now, before I hand over to David, I'm sure you see that this is going to be a very informative day. I hope we can also have some fun along the way. Speaking of fun, I do have to reference the Safe Harbor statement that is at the end of our presentation that you can find on our website. Let's go. David, over to you.
Yes. Good morning, everybody. Very happy to see you all, especially, of course, the people that made it all the way here. The people on the webcast also, a warm welcome. Now, despite the weather that you had to wrestle through, you'll see that we're all in a very good mood here because we have a lot of good and business progress to share with you, which we're very proud of. We've made very good progress on all of the KPIs that we set. We're also very excited about the plans and the targets that we have been setting for 2028. Let me get into that. What's new? First of all, what's not new is that we have a very competitive share of the value proposition.
If you look at our, let's say, our recent market cap, then our capital return yield is around 8.4%. With this plan, you will actually see that it will grow with a CAGR of 7%-8%. The dividend yield will grow with 7%-8%. That includes the share buyback of EUR 300 million. That makes the shareholder proposition even more competitive. What is new is the second bullet here. We are going to be talking about a strategic transformation that we have been running, that we started already thinking about and working on a few years ago, and that is now in full motion. Essentially, what we are doing is we are building a platform for growth. There is obviously a lot happening technologically, AI. There are a lot of opportunities that are created.
That actually works in terms of expense efficiency, in terms of customer scores, but also on growth. I'm very proud to say that we're at the forefront of these developments in the insurance industry. This transformation obviously comes on top of favorable growth trends that we see in general. The need for protection products and the awareness that people have on their needs and often the under-protection that they have continues to increase. We also see an increasing need still for long-term savings in the markets that we operate in. It's a combination of a digital and strategic transformation with underlying favorable trends that builds this platform for growth that we're working on. This is just not only just talk on trends. We've also translated it into what we think are very ambitious financial targets.
If you look at the targets, we've set a target of EUR 2.2 billion operating capital generation for 2028 and EUR 1.8 billion or more of free cash flow generation also in 2028. That both translates into a 7%-8% CAGR per share growth, both for OCG and free cash flow. What is new is that this growth will be mainly coming from Europe. It will be coming from Non-life and from Japan. Actually, that shift is different now. Europe will benefit from the need for protection products and long-term savings. As we will come back on today, Europe has some unique distribution capabilities that also enables us to get access to these customers that are currently under-protected.
Non-life will benefit from expense efficiency, trying to control claims, but also is very much at the forefront of applying AI and other tooling to continue to improve risk selection and the results. We see selected pockets of growth also in the Non-life market in the Netherlands. Finally, Japan as a third growth engine. We expect to recapture, let's say, the market share, the top three position that we used to have in this market in the corporate life space. This business mix is new because it means that in terms of OCG in 2028, the majority of OCG will actually come out of these three businesses: out of Europe, out of Non-life, and out of Japan. For free cash flow, you'll see a similar trend. Back in 2020, 75% of the free cash flow came actually out of NN Life.
While in this plan, you'll see that less than 50% of the free cash flow will come out of NN Life. This is not because NN Life is going down, because, as Annemiek will also explain, we aim to keep the free cash flow out of Life flat for the coming periods. It means it's real growth. It's a real step up of what the other entities are delivering. It kind of also confirms that Robin was talking about the IPO in 2014. At that point, of course, NN Life was very dominant, and we were often associated with kind of a runoff pattern. This also addresses, in our view, the question, are we a runoff company or not?
Because with a stable remittance out of NN Life and growing businesses leading to a 7%-8% CAGR in OCG, in free cash flow, and also therefore in expected capital return, it also addresses this point that we're really a company that will continue to grow in the markets that we operate in. Finally, since the IPO, we set a lot of targets, strategic targets, financial targets, but we've always achieved them. This was often in volatile times. I mean, it seems like ages ago, but also then there was a lot of volatility. Unexpected things were happening in the eurozone. Obviously, also COVID came along. Despite all this volatility and turmoil, we always delivered on our targets. You will see today also a team that is very committed to also deliver again on the 2028 targets.
No doubt there's going to be volatility and a lot of unexpected things that will be happening in the coming years. You will see that NN, we have built a platform that is very resilient and can deal with a lot of this volatility and unexpected events. Therefore, we're convinced that also towards 2028, we will again deliver on the target. I'm sure you'll see the same conviction in the rest of the team that you'll be meeting here today. With that claim that we always achieve our targets, it's probably good then to look at where we are briefly on our targets versus the 2025 targets that we have set. First, let's look at our strategic targets. On the customer side, we're clearly well on track. We now have seven markets that were significantly above the competition.
We have three markets where we are in line. We have no markets anymore where we're below in terms of customer satisfaction. Clearly well on track here. Keep in mind, when you're building a platform of consistent growth, it is very important that you have good retention and new business growth. This customer score is very important, of course, for this. Same with people. Increasingly, we see tight labor markets. We hunt for talent, certainly in the digital space and in certain areas. Having a high engagement and having a culture where you can tap into a larger diversity of group to be able to bring in talent is very important. Therefore, we're also very pleased to see that the engagement is very high in the company. Also on the diversity scores, we score well.
That means that we're very well positioned in a very competitive labor market, often markets that have full employment. On society, positive scores as well. One of the main targets, of course, was reducing our greenhouse footprint in 2025 with 25% of our corporate investment portfolio. It actually went down 31%. We also stepped up our investments in the energy transition to EUR 13 billion. Also on the society side, it's clear that we're well on track to deliver on the targets that we have set for the end of 2025. Now, let's look at the translation of that, of the strategic KPIs into the financials. We will deliver on the targets of 2025 that we set on the financial side as well. We already achieved the EUR 1.9 billion OCG, which was initially EUR 1.8 billion.
We later increased it to EUR 1.9 billion, and we achieved that target already last year. For free cash flow, we then set also an explicit target of EUR 1.6 billion, if you recall, which was EUR 100 million higher than initially thought. EUR 50 million we put into the dividend. And EUR 50 million we, at that point, put into an increase of the share buyback. So this EUR 100 million we passed on to you already. For this year, we also expect to make the EUR 1.6 billion free cash flow for the end of 2025. On Solvency II, despite all the market volatility and all the debates, and I'm sure we'll be having a debate on left and right some percentages, but the volatility that we have seen didn't mean that our solvency was hampered by it. It actually remained very strong. It was 194% at the end of last year.
That is basically driven by two factors, especially strong underlying business performance. We have seen that the business units continue to deliver also on their individual targets. We have also taken a lot of effective management actions. For example, we now have EUR 30 billion of longevity risk reinsured. We also settled, for example, the unit-linked case that also helped to reduce the risk profile of the company. Finally, on capital return, the strong business performance and the strong balance sheet have also translated into a very strong capital return for our shareholders. Since 2021, our dividend per share has grown 11%. This is well above the mid- to high single-digit guidance that we have given. We have increased our share buyback from EUR 250 million to EUR 300 million. Overall, next to the strategic targets, also a very strong financial track record versus the targets that we have set.
Now, probably more than enough of looking back. Let's look at where we are today. Now, today, I'm not telling you anything new. There's a lot of things happening. There's a lot of volatility. There's trade wars. There's unfortunately real wars ongoing. We see a lot of conflicts that translate into a lot of volatility. That is obviously written a lot about. I think what probably gets less attention is there's also a lot of opportunities actually in this environment. I already mentioned in the space of technology and AI, it's incredible how much advancements are being made. We're really in a phase from all this technology that is there to develop, that is being developed actually into applying it also into our businesses and then scaling it across markets. What we also see is that there is a continued need for protection products.
Most markets was already there, but there's also a higher awareness of people that governments might not always be there for them. Their employer might not always provide coverage. There is a lot of responsibility shift to individuals. That also creates an opportunity for us to provide good protection for our customers. We have also proven to be very resilient and stable in these volatile times. Let's look at our plans. We start again with the strategic targets. We look at the underlying drivers of how we manage our business that then translates into the financials. Basically, also on the strategic side, we aim to beat the competition on all fronts. That means customer experience. Again, we want to beat the competition in customer experience, but also in broker satisfaction.
As you probably know, brokers have a big influence, certainly in the SME space, on where business goes to. Also, we want to beat the competition again in engagement from our own colleagues. Again, very important to have the right talent when you're in such a complex environment where you also want to enable a lot of change. We want to reduce the footprint of our greenhouse gas, the greenhouse gas emissions of our corporate investment portfolio to 45% in 2030 and increase the investments in the energy transition to EUR 13 billion. Now, let's look at the financial targets then. The primary goal is EUR 2.2 billion OCG in 2028, which translates into a 7%-8% CAGR per share. I was saying this growth mainly comes from Europe.
Will have later on Paweł and Kuldeep speaking as examples on how they are doing this in Poland and in Romania on how they plan to achieve the growth. That should give you also a higher conviction that we will be able to deliver on this growth in the European business. On Non-life, where we see selected pockets of growth in certain product range and distribution channels and the recovery of our Japanese business. Overall, these three entities will then be more than 50% of the operating capital generation. The more than 50% will then come out of entities that are growing today, but that we also expect longer term to consistently grow. Free cash flow will then reach EUR 1.8 billion or more also by 2028, which also means a 7%-8% CAGR per share.
Now, these targets show strong growth, but it also shows an attractive shelter proposition for all of you. Maybe it's good to zoom in on, to zoom in on growth. Let's start with Europe. We have a strong European footprint. All these businesses were either built in the 1980s by NN or in the 1990s when the wall came down. We're mostly in those markets, the top three players like Greece or Romania or Hungary, we're even the number one player. In pensions, we're also often a top three player. For example, in Poland or again in Romania, we're the number one player. This business has already doubled almost since 2020. It grew to EUR 461 million last year in 2024. The strategy for the region is actually very simple and very aligned. It focuses on protection products.
It means that we aim to have selling products that are relatively high margin and they are capital light. There are two key items for growth in Europe that are driving the growth. One is that we've been, and we've actually been seeing this for a long time, there is a clear underpenetration in the market for protection products. On average, in a market, in an OECD market, you see 8%-9% is spent on insurance premium. A market like this is also in that space. In markets like Poland, like in Romania and Hungary, it's actually much lower. It's often more around 2%, 3%, or sometimes even below 2%. That is one element. What adds to that is that because of all the uncertainty in the world, we see an increased awareness. COVID certainly played a role.
Customers and people are becoming more and more aware, like, what if something happens to me? What if something happens to my family? How am I protected? That is one macroeconomic trend. The reality is also, as I was saying, if you look 10 years ago, this underpenetration was also there. By itself, it's clear that this will not lead to customers buying a lot more protection products. That is why the second element of growth is so important. This is our unique distribution capabilities. Very few people wake up in the morning and think, hey, I'm going to buy a protection product. You still need actually distribution chance to sit with a customer, and not that these products are so complex, but usually the situation of the customer is complex.
They need to figure out what gaps do they have, what do they have already covered, maybe via the government, maybe via their employer, maybe they have some protection. What gaps do they have, if anything, and how can I fill these gaps? That is where you need strong distribution channels that are able to have access to these customers and can help and advise them on the type of protection products that they need. You see it on the right-hand side. We have a very multi-distribution platform, as we say it. What we mean with that is that you see per market differences, but you also see over time different trends. We work with over 10 banks. Brokers have been growing quite rapidly. Tied a gent has been the largest channel more recently. These trends differ.
Pre-COVID, banks were actually a very large channel. Then COVID hit. Banks essentially stopped selling mortgages. That meant that also the amount of credit insurance that we were selling came down significantly. You did not really notice that in the growth numbers of Europe because tied agents were really able to step up during that period. The fact that you have multiple distribution channels that can diversify the risk profile is a very important feature of the European business that we have. It actually led to a 44% increase in growth also if you look from 2020 to 2024, which is a very strong number. It means that for Europe, we are very well positioned. We had a target of EUR 450 million, and we are now increasing the target to EUR 600 million, which obviously makes Europe a very significant part of our company.
Now, let's look at the next growth engine, which is Non-life. When we were thinking about the Capital Markets Day, you can also go back in time. Quite a few actually, and we're also here. In 2017, we had a Capital Markets Day, and I was talking about the Non-life business. I said, we're going to aim for a combined ratio of 95%-97%. People were all kind of looking at me. Next day, same day, you guys were all writing reports on that's not going to happen. I mean, this market has gone under a tremendous transition. I mean, even more than 10 years ago, even the DNB was writing premiums need to go up. This is a badly priced market. Since then, a lot happened. Obviously, pricing discipline has tremendously increased.
Consolidation was a big driver of that. We took over, obviously, VIVAT. We took over Delta Lloyd. That led to a situation where three players have, let's say, roughly 2/3 of the market. That is why we've been now able to consistently produce good results. We've always focused on saying we need to get to margin over volume, meaning 91%-93% as the target that we maintain. We do see in that market also selected pockets of growth. If you look at our positioning, we're overweight in fire, meaning underweight in motor. In the disability space, we're overweight in sickness, which is the shorter duration disability products. Those are actually also the products where we see more growth opportunities. Both in fire and in sickness, we see growth opportunities to further grow.
If you look at the Non-life distribution profile, I would say that's really unique. I mean, if you think about it, the Netherlands is traditionally very much a broker market. NN Non-life is the number one in broker business, but also the number one in customer satisfaction of brokers. Trust me, that is a big driver of your commercial success. Non-life is also the number one in mandated agents, so mandated broker agents. Also there, the number one in customer satisfaction. Out of the five largest banks that we have here in the Netherlands, four are exclusively tied also to us.
If you go to an ABN AMRO branch, or you go to an ING branch, or you go to an SNS branch, and you get an SNS or an ABN AMRO product, it's actually an NN underwritten and administered product from us. Finally, with OHRA, we have a strong direct brand. The distribution profile of Non-life is very strong. In terms of channels, we expect mostly from the bank and direct to expect a bit more growth, while we expect the broker just to continue to develop well as they have been doing. There's some upside on the direct side and with our banking partners. This is combined with strong expense discipline. As I was saying earlier, and Rob was saying, Non-life is really at the forefront of some of the developments in AI.
Later on, Tjerrie and I will give some examples of how we actually already have quite a few examples now live. It also shows how much potential there is after that. In conclusion, Non-life has been doing well. We had a target of EUR 325 million initially for up to 2025, and we're increasing this now to EUR 475 million for the Non-life business in 2028. Japan, our third growth engine. If you take a step back from Japan, and Marius is also here, so you can have a good conversation with him as well this afternoon. If you take a step back, the overall outlook for Japan is positive. We see some increased interest rates. We see the economic outlook is positive. Inflation is a bit better under control. This is especially also important for the SME market.
The SMEs is by far the largest part of the economy, but it's also a bit more cyclical on average, therefore. Now, we've built since 1986, we've been in this market, and we've built a really strong franchise focus on the SME market in what we call the corporate life market. Now, Japan, of course, is the third largest insurance market in the world. You can imagine that also therefore the corporate life market is a very, very large market by itself that we operate in. We used to have a top three position there. The business improvement order came, and that meant that our sales was roughly cut in half. We're now in a lower position. Since then, we made a lot of progress. A lot of change was implemented and a lot of reporting.
We are now in the final stages of the business improvement plan. We have filed all the documentation to the regulator, and we are now essentially waiting for the regulator to come back with their verdict. We do believe that we are in the final stages of this business improvement plan that we have been working on for the last few years. It also means that in this plan, we assume that sales will go back up. What is interesting, I think, there is that traditionally the Japanese market was selling protection products and protection products that had what we call a short-term cash value. Short-term meaning probably around five years. Now, that five-year product acquiescently came under pressure from the regulator, the business improvement order that we received. We were not the only one. There are a couple more that were under this business improvement order.
Now, and I spoke to most of you earlier on this. We've always been saying we do believe that a long-term corporate life market will emerge. That is actually happening. If you look at the first quarter, the long-term COLI market actually grew with 17%, while the short-term COLI market actually came down with 8%. It proves for us that really the long-term market is now developing. We will continue to have protection products. We will for sure have the long-term corporate life products. There is probably some version of short-term products that will remain. We were also very pleased that we did get an approval of the first long-term COLI product already in April. We started already selling. The first results are very encouraging. We've seen already the sales coming in in April, also the sales that we've been seeing coming in in May.
It confirms for us it is very realistic that we will start growing back this business. It will take time. Ultimately, we do expect by 2028 that we will be back at the VNB levels that we had in 2022, and probably in a bit in a different composition with more long-term products and more protection products. We do expect to reclaim our leading position of top three player in Japan. This also translates into a higher OCG. The target will go from EUR 125 million where it is today to EUR 160 million in the end of 2028. As I was saying earlier, when we achieve all of this, it also means that these three entities will be more than 50% of OCG. You can see that here on the pie chart. Orange is NN Life.
It used to be more than 50%, and now these growth entities will be more than 50%. You see a similar trend in free cash flow, where NN Life was 75% and will be now by itself below 50%. Again, for NN Life, we actually project stable cash flows. In fact, we even talk about 2040 and that we can keep the remittances flat up to 2040, but that is later on for Annemiek to cover. Now, apart from this shift in segment, we also think there is another shift that is important and that we like. Over time, when you see the shift, it also means we become a bit less dependent on spread or on market risk. Actually, the amount of fee business and technical margin is increasing.
Non-life, obviously, fee business and especially technical underwriting, but Europe with protection is technical margin, but also the growth of the pension book is actually fee business. The shift is not just between segments, so you see more growth, but we think also a healthy shift between a reduction of spread versus an increase of technical underwriting and fee business. Now, clearly, when we achieve all of this, we have capital to allocate. We will remain very disciplined in our capital allocation. Obviously, our first priority is to make sure our units are well capitalized and they can have the capital to grow. There are strict criteria on the returns that they need to make. For the rest, we will give them all the capital that they need to actually grow against these attractive returns.
Now, clearly, you still have then capital for dividends and share buyback. I spoke about the growth in dividend per share that we're projecting also on the back of the growth in OCG and free cash flow. We will still have excess capital, where we remain saying we will also give that back to shareholders unless we see value creating opportunities. It's important to keep in mind that when we talk about giving excess capital back to shareholders, we aim to do this in incremental steps and in a recurring way. We do this once the Solvency II ratio is sustainably above 200%, which is today with the 195%, obviously not the case yet. Now, finally, since the IPO, if you look at the capital return promises that we've done, we have fulfilled always our capital return promises. We've given EUR 10 billion back to shareholders.
I was saying earlier, we project a 7%-8% growth of capital return per share towards 2028. Keep in mind that this already comes from a very attractive free cash flow yield of around 11.1% in this period. That brings me to the key takeaways of this first part of the presentation. We have set very clear strategic and financial targets for the group. Our strategic transformation is accelerating. Information technology, artificial intelligence play a big role in further supporting, let's say, the growth that we see macroeconomically in the business. We will come back on that in the next presentation. We aim to achieve EUR 2.2 billion OCG, or 7%-8% average growth per share. Free cash flow, EUR 1.8 billion or above, also aiming at a 7%-8% CAGR growth, fueled by higher remittances from Non-life, from Japan, and from Europe, of course.
All of this is then also a testimony of our commitment to provide stable, attractive capital returns for our shareholders. That means for our shareholders, we expect also 7%-8% DPS annual growth. Again, this comes from an attractive yield already of 8.4% versus a recent market cap. This concludes the overall part of my presentation. I'm sure you'll have lots of questions if I see the amount of notes that you guys are making, but you need to hold off for that for a while because we still have a couple more presentations before we go into Q&A. Let's talk about the Future Ready and the digital transformation that we're currently going through.
We thought it was good to do a deep dive on that to give you more background on actually how we're achieving the things that we're planning on doing and why we're therefore also convinced that this will help to get us to the targets that we have set for 2028. I will do this presentation together with Tjerrie Smit, our Chief Analytics Officer, and who has been instrumental with us and has been bothering me since 2015 that I should pay more attention to this. He was fully right, and it took me some time to learn that. You will see that Jerry has been instrumental in this journey of 10 years of how we have been building our capabilities. Now, let's start with something that you already know, which is agentic AI. Gen AI is clearly the latest trend. I mean, it's the hype.
Everybody's talking about it, and it became very mainstream, of course. Now, if you look at our industry, the insurance industry, it is very, very suitable for applications of AI. The reason is it's complex. Mostly because of the complexity, the industry hasn't been very successful to address the expense challenges that we see in the industry. Second, in our business model, there's lots of risk selection ongoing. There's lots of claim handling ongoing, and both are also very suitable for AI applications. Finally, we have a lot of customer contact. Already we see in the space of contact and bots and avatars, and there's a lot of upside in further digitalizing our customer contact. Clearly, the insurance industry is very suitable for all the technological advancements that are being made. You haven't heard me speak much about the things.
When we had blockchain coming out and the Internet of Things and Lean Six Sigma, there's obviously been a lot of trends already ongoing for a long time, but we haven't spoken much about it. The reason was always that whenever these elements came out, often it felt like there was some basic intelligence we're missing. We were programming things where we were digitalizing certain processes, and then something would change, and the system would just continue to do what it was programmed to do, and it wouldn't be able to adapt. Then any colleague of us would look at the digitalization and say, "Wait a minute, this is not right anymore." Anybody would see that. The issue was that the system itself could never do that.
I think with the arrival of AI, it created a basic intelligence function, and it could adapt things and adjust things that was not possible quite a while ago. That has been a big change for us when we saw that, and Tjerrie has been explaining that to us. That means also that since then, we've really been investing. Now I'm talking about since 2015, because since 2015, AI started to become more commercially applicable. In 2022, when Satya Nadella and Sam Altman came out with ChatGPT, it was therefore also not for us a coincidence that we were one of the 16 brands, one of the 16 companies that they mentioned and they put on the screen, because we had already then for years been actively engaging with OpenAI on large language models, including eventually what led to the launch of ChatGPT.
We've made a lot of mistakes along the way. I mean, it's not an easy journey. We made a lot of mistakes. We had a lot of things to fix, redo. I'm very proud that not only then at 2022, but still today, we're very much at the forefront of this. The reason why I think we're at the forefront is we're not just investing. We're not just having a center of excellence that everybody has. We're not just having proof of concepts, but we actually have use cases that are live and that we're scaling. It started all with automatic call logging that was developed and then was copied into Spain, into Japan. By now, we have multiple cases that are actually live and that we're scaling.
Instead of just investing and building proof of concept, we're actually in a phase of already scaling AI and even agentic AI in our company. Now, AI doesn't mean much if you don't put it into a business context. I think one of the challenges you often have is none of us really grew up in this world. It unfortunately started with educating ourselves. We went through an extensive training ourselves, including myself, with the board, with senior managers, with the country managers. We went through a whole program of over a year at MIT. This wasn't your, let's do a week at MIT and everybody gets a diploma, and then you put it on LinkedIn. This was real suffering. We had online, we had obviously the business cases, all the technological developments.
We started with educating ourselves, which already we did a few years ago. We have been continuing to do that. Even recently, we went with a group of 30 people to Seattle to see the latest developments now in agentic AI. One is the education, but we ended up putting it into a business context, which is what we call the NN Future Ready program. Of course, any consultant or business school you deal with, you always get four quadrants. I do not know. Here you go. We have four quadrants. The reality is that most insurance companies, and we were no exception, the backend is very much what we call silos and spaghettis. That means you have lots of applications. I mean, we talk about we are transferring it from two applications to one, but there is a whole world of applications around these core systems.
We have been running a program for years now on IT simplification and further standardizing, or as we call it, moving here to the right, meaning we are transforming and making the company more industrialized, simplifying the landscape, having very clear data governance, which is crucial to do anything in this space. Ultimately, with the goal that you end up in what we call a future ready state, which is obviously at the right top, where you not only have a more standardized data readily available, but you also see therefore much higher customer scores. Probably even more important, what you do not see here on the slide, much higher employee engagement scores. It is just an easier, more fun company to work in if data is readily available, systems are standardized, and when you have ideas, you can immediately apply it yourself. That is the journey that we have been on.
It is not just around the business case, but it is also around that you build a company that is much better prepared for whatever the world is going to throw at us, because we all know that the world is fairly unpredictable. In order to get your house much better in order, prepare you also much better for an unpredictable future. Now, I am not going to bore you with all the details of the program. Trust me, we do. We have steer codes, and we steer on all of these. I know most of you also love numbers. Let me give you some numbers on this program that we are doing. We started this program officially in 2024, and we invest EUR 450 million over the period between 2024 and 2027. We expect a benefit of EUR 200 million recurring out of this.
That's a remarkable short payback period of around two years. Now, these numbers obviously are incorporated into our targets. When I was earlier talking about our 2025 targets or the other CAGR targets that I spoke about, obviously these are incorporated in the targets that we have given. In fact, if you look at capital, the majority of these investments have already been taken in 2024. That means that only for 2025 and for 2026, we expect at the holding level a EUR 50 million impact on free cash flow for 2025 and for 2026. It does not jeopardize our EUR 1.6 billion free cash flow target that we set for 2025. If you look at the benefits, as I said, there is EUR 200 million in benefits. EUR 180 million is expenses, and around EUR 20 million is growth. This is actually part of the journey you go through.
When you start with AI, you think of all these cool cross-sell things and all these cool next best actions. The reality is, in a complex industry like ours, there's a lot to gain from just simplifying and reducing your cost and becoming more competitive. I do expect over time, if you go long term, that that mix will shift a bit. For now, the vast majority is expense savings, and then there's EUR 20 million of growth in there. This might all sound a bit high level, so let me give you two examples of what we actually have live and what we're scaling across markets. The first one is our digital lead program in Europe. As is often the case, this is a collection of smaller AI use cases. This one is particularly geared to tied agents.
Tied agent channels always have a couple of challenges, which is usually the retention of agents. How do you train them, but how do you also keep them on board after they've sold to their friends and family? How do you keep them in a, how do you keep them that they make enough money to consistently keep selling, especially in the protection space? Protection typically are relatively cheap, small ticket size products. You can also imagine that therefore the commission is lower, and it means you need to sell a lot more than if you're selling big endowments or big unit-linked savings type of products. What do we do? One, and Poland is actually a very good example. We generate a lot of online content on protection.
That means that if customers are online, there's a good chance that they'll end up in an NN environment because we generate a lot of content around protection products. That means there's a good chance that they will end up in our environment after that. One of the other things we do then, based on everything that we know from customers, we link the profile to the right profile of the agent. To be honest, when I heard it, I thought a bit, is that a big deal? The people have been explaining to me, they said, well, this agent is going to be serving you for the coming years, and it matters. Finally, I understood it when they said, so we would never give you an agent that is not also able to talk about football. Then I thought, okay, now I understand.
It is really important that you get a good connection between the customer and, let's say, the agent. Next, we are in Europe, which means there is an enormous amount of regulation. I mean, you would not believe the amount of documents that an agent or anybody has to sell when he is selling a product, has to show to the customer, talk them through, explain, get them to sign. What we have actually built is our internal ChatGPT environment. That means that, in a way, this agent has now the smartest, nerdiest colleague always with him that can help him with any of this documentation, any questions, any things that he has. He can type it in, and ChatGPT, the NN version of it, will provide him with the right answers and also make sure that everything is done in a compliant and in the right way.
It creates an enormous amount of efficiency. To be honest, the average tied agent, not so interested in compliancy and in documentation, is often much more interested in engaging with people. It is a tremendous help. After the conversation between the agent and the customer, another version of ChatGPT actually analyzes the conversation. It will tell the agent, this is where you deviated from the script. These are the opportunities that you might have missed. This is potentially a next best action for this customer. It will provide statistics on it. Jerry will later on show a bit more detail on it. This is very, very valuable. When you are dealing with thousands of agents, you need to continue to train them. The efficiency needs to go up.
A version like this that is very personalized will then do a tremendous job in actually helping to train and improve this agent. Finally, based on all the activities, another AI bot will tell them, this is the amount of commission that you're going to be getting in the next few years. You actually, versus others, spend a lot of time on retention maybe, or you spend a lot of time on sales, or this is how you can optimize even to further increase your income. This is very important because a lot of agents leave at some point because they think, I'm not going to make enough money on this, and I'm going to do something else. There are a lot of these AI use cases just to support a tight agent channel. Now, let me share some results.
We've already seen that the number of active agents, since we've been doing this in the last few years, has grown with 26%. We have 1,400 more active agents now. Active agents means that they sell two products per week or more. You often see statistics on tight agent channels, but actually active agents are usually more important than the number of tight agents that you have. Another data point, which I think is very impressive, is we now see that within Europe, with the tied agents, 36% of their sales actually comes out of this digital lead program. What is very interesting about this 36% is that markets like Romania and Poland, where this started, they're actually above 50%, while Greece is around 20%. Greece obviously was very busy integrating MedLife, and so they've been focusing on other things.
It shows the potential when you're scaling this across markets that there's still a lot of room for units, not only for Poland and Romania to further increase, but also the other ones to catch up and based on the learnings that they have. That's one example. Now, let me cover another short example in the Non-life space, and this is around claim handling. Unfortunately, I personally had to test this one, or actually my daughter. She borrowed my car, and it looked a bit different after I got back. Wasn't serious, but what was very cool is that you make a picture, you send it in, and very quickly in that process, you immediately get a response saying, this is how much it will cost. This is where you can repair it, and if you repair it there, you won't have a deductible.
This is the impact on your premium. You can immediately, as a customer, make a trade-off between am I going to pay this myself, or am I going to have NN pay this? Of course, it will come out of your premium, your premium will increase, and as a customer, you can immediately make that trade-off. What I thought was even cooler is that if you have, and this is now for windows, if you break a window of your car, again, you make a picture, and we now have, and this is actually the first version of agentic AI that we have live, meaning that an AI agent starts working on it. One AI agent starts analyzing the picture. You basically go from unstructured data to structured data and will analyze, yes, this window is broken.
Another AI agent can actually make an assessment on what is the claim amount. Is this a reasonable amount? Is it too high or too low to fix this? There's another AI agent that will actually check the policy conditions, will check is this covered or not, check the license plate, if there's something else suspicious. Now, if all of that happens in parallel, it means effectively that we've now had many cases where within five minutes, the claim is paid out. In fact, one of the brokers actually called and said, you guys made a mistake because I just filed the claim and the money is in the account. That means that if there's no red flags, it means immediately the money is paid out. Within a few minutes, actually, the payment was in the account of the customer.
Now, you can imagine this is now for window shield. You can imagine that you can do this for other types of damages on the car. You can do this for windows of buildings, very large fire insurance for homes. You can imagine that we're only at the beginning of this, and you can continue to scale this across many other types of damages because, in a way, the process is often very simple. These are just two examples that we believe will help transform our business and will support also the targets that we have given. Now, there's a whole world, obviously, under there that is a lot more complex than I make it to be. Therefore, I think it's very good that Jerry will also take you along on the journey. Terrie.
Thank you, David, and good morning, everyone. I started doing full-time AI back in 2015. Back then, it felt like often I was the only one pushing for what we today call AI, completely different than it is today. Today, I'm on stage mostly two, three times a week presenting on AI. Last week, I was in a professional studio where I recorded a digital copy out of myself, an avatar that I can use in the future to present on behalf of me. A few years ago, I could not have imagined that I would be doing something like this, let alone that the quality of the AI technology that we are using today here at NN would be this good. This is because the developments in AI technology, they are really at a record pace. David already said it's almost impossible to keep up with it.
This is actually good for companies like us, companies that are early adapters and operating at the forefront of this new AI technology. Today, I will show you how we execute AI at scale here at NN. For our employees, we quickly opened up all kinds of GenAI tools to make sure that they can work in the most efficient way, but also so that they can experiment safely with our GenAI tools. For instance, we opened up our NN ChatGPT playground in less than three months after the release of ChatGPT. We have now more than 9,000 employees working in this environment today. For efficient coding, we have more than 1,300 engineers working with our GitHub Copilot environment so that they can be more efficient in writing code. We are making significant progress in implementing GenAI technology.
We are implementing AI not only in pilots and POCs, but we are implementing it at the core of our business processes. We can see clear proof points for a fundamental business transformation with the help of AI. Our focus is to scale on scalable and successive AI use cases. We have AI use cases that contribute both to process efficiency as well as use cases that contribute to our top-line commercial growth. Before I dive into the use case applications, I want to share with you our AI setup. We have adopted a hybrid setup where we have a center of excellence here in the Netherlands and Czech Republic, complemented by many, many local data scientists, embedded as close as possible to the business.
This is really important because a long time ago, we learned that having data scientists in the business is crucial for a successful implementation. Designing AI use cases together with the business is the most effective strategy for us. What we want to do is we want to develop standards centrally, develop reusable components that then the local data science teams can apply in their business. This approach not only helps us to scale AI, but it also makes sure that whatever we put out there is safe and reliable. Like I said, AI is developing at a record pace, and it's getting better really fast. I think, David, you also touched on that point. Everyone can build a pilot or a proof of concept. Executing AI in production 24/7 is really a different game. Doing this at scale, I think, is even harder.
This is because this requires different skills, but it also requires experience. This is where our early investment in AI really pays off. We made a lot of mistakes in the beginning, but we also learned how to play this game. I think everyone needs to go through the same learning curve. Like I said, our focus is on scalable and successive use cases. At scale, use cases are kind of like railroad tracks. It requires time and craftsmanship to build it. Once you have the tracks there, you can run many trains on it. Our focus is therefore to copy and paste the really good use cases throughout the companies of NN Group and to benefit from our economies of scale. Now, let's dive into some exciting use cases.
I selected these use cases because they are scalable throughout NN Group, and they act at the core of our insurance processes. First, I will share with you how the personal expert assistants are improving our customer experience and also improve our efficiency. Secondly, I will show you how commercial AI assistants are helping our sales agents to become better at converting leads. Third, we will see how we enable our customers and partners to interact with us at any time with AI-powered chat that actually works. Let's have a look at the first one. This is about the personal expert assistant. This is a RAG technology solution. Of course, everyone knows this stands for Retrieval Augmented Generation. This means that we are grounding our own knowledge documents to a large language model.
Basically, this means that our employees can talk to ChatGPT, but for the answers, we only use our own context and knowledge. The results are really, really good. Applying the AI expert assistant in one of our bancassurance channels here in the Netherlands resulted in an almost instant uplift of the contact center performance. This means a 25-second less call duration, 8% less hold time, and even a 30% fewer transfers to senior employees. If you are familiar with contact center statistics, I am sure you will appreciate these numbers. They are huge. Let's see how this looks in practice. On the screen here, you can see that I asked the expert assistant if a damage caused by flying a drone is covered under the liability insurance.
It will clearly state to me that the damage is covered in case the drone is less than 25 kilos in weight, if it is flown in the EU, and no other liability policy will cover it. At the bottom, you can see the links, deep linking to the original knowledge documents that were used to create this answer. This is because maybe I want to check or in case I need more information. You could ask, why does this matter? The insurance industry is a very knowledge-intense industry. Policies, legislation, procedures, they change all the time. We have literally thousands and thousands of knowledge workers where we can apply this. With the personal expert assistant, we put the right expertise at the fingertips of not only the novice employees, but also our experts.
I want you to think of the most brightest, knowledgeable person in a department. This is like having that person helping you all day. We do this at scale. We have this operational now in 35 implementations in production, not only for our employees, but also for our tight agent network. In NN International, here, we use the same technology. By the end of the year, we expect to have more than 100 of these personal expert assistants to be live in production. Having such an expert helps us to increase the customer experience, but it also improves consistency and quality. Next to that, it's more efficient. I think this is a perfect example on how we apply AI use cases at scale throughout the companies of NN Group. The next example I want to show you is about the AI commercial assistant.
Like I said, we have use cases that contribute to process efficiency, but also use cases that contribute to our top-line commercial growth. This example is in that last category. It is from NN Romania, and it is all about sales lead conversions. Here, we use speech-to-text technology together with LLMs, that is, large language models, so that with that, we can help our agents to become better at converting their leads. Let's see how it works. As an agent, I can see clearly an overview of all the calls that I did, together with the overall performance and statistics. Here, you can see, for instance, the speak time of myself versus the speak time of the customer. More importantly, I also get an insight on how well I followed the call structure.
You can see that in cases where I followed the call structure, I have a 4.3 better conversion rate than in cases where I did not follow the call structure. This may look very simple on the screen, but let me assure you, there is a lot of AI technology driving this in the back. By the way, it also proves that following our call structure actually works, fact-based. Imagine if we could enable all of our 9,000 agents with this technology. This is exactly what we are doing. We are rolling out to other countries in NN Group. On the technical side, we are also adding more features like next best actions and real-time agent assists.
The next topic that I want to talk to you about is AI agents, because you can't have a presentation these days on AI without speaking about AI agents, right? This is because AI agents are really powering our AI transformation. AI agents or agentic are the biggest thing after the introduction of ChatGPT. For companies like us, AI agents are more impactful than ChatGPT itself. Let me explain in the shortest, most easiest way possible what are AI agents. AI agents can perform tasks and act on your behalf. Basically, AI that can actually do stuff. AI agents or agentic are a good fit in situations where you have complex decision-making, where you have difficult-to-maintain business rules, or in situations where you rely heavy on unstructured data.
This is why agents are such a big deal for us, because in most core insurance processes, such as claims, underwriting, but also mortgages, we rely heavily on unstructured data. It requires multiple documents like photos, emails, scans, and other inputs that we need to interpret and validate to come to a decision for a customer. Frequently, those documents are incomplete or not delivered in the right way. The problem is still customers expect a fast and frictionless experience from any digital company nowadays. We think this can all be enabled by a combination of human experts and AI agents. A good example was already shared by David on our windscreen fully automatic claim handling, where we are able to validate and come to an automated interpretation and payout for most of the claims.
This afternoon, in the workshop of Non-life, you will see a demo of that use case. Next, let's move over to agentic and agentic chatbots, the next generation of chatbots. This is another example where we apply AI agents in the core of our processes. Let me also be really clear about chatbots. They do not have a good reputation. People do not like traditional chatbots. Often, the most frequently asked question is, can I speak to a human? Right? You all know it. This is really because traditional chatbots can only answer 10, maybe 20 most frequently asked questions. They are terrible at answering long-tail questions. This is where agentic or AI chatbots are really different.
A good agentic chatbot requires three things: a good knowledge base to ground the data, a good retrieval engine to interact with the customer, and good quality questions. Here's where the difficulty starts, because the first two, of course, we control. A good quality question is asked by a customer. We, as humans, are not used to being very complete and providing a lot of context for questions. Let's look at how we fix this with our new chatbot, Sonnet. Here you can see on the screen that I'm using the chatbot to ask a question about my pension. The first question I ask is very short and very vague. I do not provide a lot of information. Combine pensions.
As you can see, our chatbot will then start a conversation with me to get more information about the topic that I'm asking a question about. I respond, but still my response is very short and not too clear. After that goes on for a few rounds, I provide information that I have already a pension plan with NN. Sonnet then asks me, do you want to combine it? As you can see, the conversation that Sonnet starts is started in a very natural way with me. In only a few steps, Sonnet collects all the additional context that is needed to answer this question really well. The great thing is that none of the questions or the answers were pre-programmed in our chatbot. All of this is done by the help of AI agents.
This is now live in NN Pensions here in the Netherlands, and we will roll out to other units in 2025. This is important for us because this new agentic chatbot is 60% better at answering long-tail questions. That is a big deal. We are confident that we can scale this to other parts of NN, basically to all of the contact centers that we operate. Since the start of our future-ready strategy, we can already see an 11% shift towards digital channels, like chat, for instance. One more thing. What if we could apply even more functionality to such an agentic chatbot? What if we could add even more AI agents that can actually do stuff for our customers, perform actions on behalf of them, not only answering questions? What if we could add voice for that matter?
What if we could talk to Sonnet? That is exactly what we are working on. Today, I am able to give you a live demonstration of a look into the future of our voice bots. Are you there?
Yes, I am here. How can I assist you today?
I am here together with David, and he told me that we are going to take a trip. I would like to know, can you help me with a quotation for travel insurance, please?
Of course. I would be happy to help with a travel insurance quotation. Could you please provide me with a few details to get started? What is the desired start date for the travel insurance?
Sorry to interrupt, but there are a lot of people waiting here for me, and I am kind of in a hurry, and I already know what you are going to ask. We are leaving June 1st, coming back June 7th, traveling with two, and we're going to Spain.
Thank you for the information. Based on your details, here is a quick quote for your travel insurance. Start date, June 1st, and...
Oh, sorry, but David is just signaling me. He wants to stay longer, the weekend also. Can you add two more days for me?
Certainly. Extending your trip by two days would change your return date to June 9th. The new dates are: start date, June 1st, end date, June 9th. For the additional two days, the cost will be EUR 15 per person, totaling an extra EUR 30 for both of you. The updated total will be EUR 60. Would you like to proceed with this updated insurance?
EUR 60. I think I have to consult with David because this is definitely above my pay grade, right?
I understand. It's always good to discuss these details. Feel free to take your time and let me know if you have any more questions or if you're ready to proceed. I'm here to help.
Thank you. You can see how it works. I'm really excited about this because I think this is really the future of how chatbots will work. You could see that, or I should say, you could hear that I interrupted this chatbot on multiple occasions, but it did not break, and it felt quite natural. It was almost like talking to a human. We believe in a future where the customer can choose whether he or she wants to use a human channel or one of our digital channels. For instance, imagine that you just came out of a night shift, then you probably want to use one of our digital channels.
For more high-impact situations, you probably want to talk to a human. As a customer, we think you should make that decision and choose for yourself. I showed you four examples on how we are applying AI at scale here at NN in the core of our insurance processes. Today, I showed you how we translate the fast-moving developments in AI into our future-ready strategy. I hope I was able to show you some concrete examples on what future-ready is for us. With that, I would like to give the floor to David.
Yeah, thanks, Tjerrie. Fun that we're going to Spain, actually. Yeah, yeah. Let me be clear. The technology is there. In our view, this is really a management problem, right? The technology will continue to evolve, more and more will be possible, but the real challenge is a management problem on how do you roll this out in larger organizations. That is why we have set very clear KPIs. I mentioned already the EUR 200 million benefit that we are aiming to get, but we have also underlying KPIs that we set here. For example, that we want 300 AI cases live in 2028, and currently we are still clearly below 200. Our digital interactions, and Tjerrie was just showing today, we are roughly at 60%. We want to go to 80%.
Often the last part is the hardest part because the more complex parts today are not done in a digital way. As you can see, there is a lot of opportunity to do this. I am personally convinced that over time, more and more people will actually prefer the digital solution, but we will see how customers deal with that. We want to go to 80%. The last, as an example of an important KPI, is the digital leads program that I spoke about earlier from the tied agents in Europe. Today, at 36%, and we want to get on average to 50%. In the example of Greece and all the other entities, we think there is a good step up to be made in terms of further professionalizing our tied agents, which will then tremendously support the growth profile.
Obviously, the growth of Europe in OCG is very much driven by new business contribution coming out of these types of programs. Obviously, as I was saying, this is a management problem, and we spend a lot of time on this to support, have the technologies to support our overall targets. That means that we also continue to train. Currently, we're training another 1,500 managers within NN. After training senior management, we're now in the process of training another 1,500 managers, and we've committed to train till the last employee within NN in this digital area because the only way we can get this to work is that we don't have a center of expertise coming up with things, but there's a lot of push coming out of the unit with new ideas, new concepts that we can continue then to decide, implement, and scale.
That's why the education part is really, really important. In conclusion, we've spoken about the underlying growth of our business. Obviously, in Europe, through the protection gap that we see in our unique distribution capabilities, the growth of Non-life that we expect to come, and the recovery of obviously Japan. Now, by itself, all of these trends will not bring the 7%-8% CAGR that we have been talking about because we also need to offset the slowly run-off of the life company, which fortunately is now only at 2%, but the combination still requires more. That is why with these Future Ready programs, we're convinced that we will get to the 7%-8% as a group, not only for now, but we also believe that there will be benefits after 2028.
Now, if you look at the last period, in general, we've seen that most stakeholders, and not just shareholders, but stakeholders have been satisfied with the overall progress of NN. What that means is that when that is happening, that's exactly also the time you need to transform your business. When things are maybe a bit more calm outside, this is exactly the time to do a big internal transformation that prepares you not only for everything that is coming, but certainly also will help you prepare for everything that we do not know that is coming. That is exactly what we have been doing in the past few years, and this should give you an insight into some of the things we have been doing underlying to make sure that you see that we will deliver on these targets that we have set.
With that, I think you guys earned a coffee break. We will see you back in 15 minutes.
I like the dirt off your name with the shirt off my back. I thought that you'd do the same, but you didn't do that. Said I'm the one who's wanted for all the fights you started. You knew the house was burning down. I had to get out. You laid your saints and sinners. They're fed and not spoken of. You knew the house was burning down. Look at you now. Look at you now. How do you sleep at night? No one to hide behind. Betrayed every alibi you had. You had every chance to make a mistake. You got drunk on betterness, and you still claim that you didn't say it said that you said I'm the one who's wanted for all the fights you started. You knew the house was burning down. I had to get out.
You laid your saints and sinners. They're fed and not spoken of. You knew the house was burning down. Look at you now. You prayed for me. You were praying for my downfall. You were taking a break from me. We were sharing the same for all. You used to tell me you'd pray for me. You were praying for my downfall. You were taking a break from me. We were sharing the same for all. Thank you. Said I'm the one who's wanted for all the fights you started. You knew the house was burning down. I had to get out. You laid your saints and sinners. They're fed and not spoken of. You knew the house was burning down. You used to tell me you'd pray for me.
You were praying for my downfall. You were taking a break from me. We were sharing the same for all. You used to tell me you'd pray for me. You were praying for my downfall. You were taking a break from me. We were sharing the same for all. Thank you. Let's get one thing straight now, baby. Tell me, tell me if you love me or not. Love me or not. Love me or not. I'm at the house one, yeah, I'm a lucky and not. Lucky and not. Lucky and not. You gotta tell me if you love me or not. Love me or not. Love me or not. I've been wishing for you, I'm a lucky and not. Lucky and not. Lucky and not. You know what I need, I'll be getting up later.
Y ou gotta tell me if you love me or not. Love me or not. Love me or not. I'm at the house one, yeah, I'm a lucky and not. Lucky and not. Lucky and not. You gotta tell me if you love me or not. Love me or not. I'm wishing for you, I'm a lucky and not. I've been wishing for you. Trying to put one love is through. I've been wishing for you. Trying to put one love is through. What love is do? Are we too grown for games? Are we too grown to play around? Young enough to chase, but old enough to know better. Are we too grown to change? Are we too grown to mess around? Ooh, and I can't wait forever, baby. Both of us should know better. I've been wishing for you.
Trying to put one love is through. I've been wishing for you. Trying to put one love is through. I've been wishing for you. Trying to put one love is through. I've been wishing for you. Trying to put one love is through. I've been wishing for you. Trying to put one love is through. Maybe it's me, or maybe it's you. Maybe I never learned my lesson. Someone to see, someone to soothe, someone to answer all your questions. I don't know what to tell you. Maybe I never knew. Not trying to make you jealous, but it looks so good on you. Been stuck on elevators and airports and empty rooms. Been stuck here waiting for you. Crying out loud, I'm calling your name. You wear me out, but I love you the same. Crying out loud, your heart is a maze.
You're burning your mouth, but I'm singing your praises. All of your clothes all over the floor. All of it happened in a hurry. Show me the rose, then you show me the door. Why don't you show me why you're so worried? You're sweating off your makeup. You're running every crack. What's gonna help you wake up? Hold up your sleep at night. Sometimes I see the future. Sometimes I'm not so bright. Sometimes you're looking for a fight. For crying out loud, I'm calling your name. You're wearing me out, but I love you the same. For crying out loud, your heart is a maze. You're burning your mouth, but I'm singing your praises. Crying out loud, I'm calling your name. You're wearing me out, but I love you the same. For crying out loud, your heart is a maze.
You're burning your mouth, but I'm singing your praises. Watch what you say when you're talking to me. Needed all the heart of me. Parking away with the kids on the cheek and a bruise on you. Crying out loud. Love, let's talk about love. Is it anything and everything you offer? I do it for your heart, your heart. I know it for your heart, your heart. 'Cause maybe the night that I'm with you is my light, make me know. Only stars are closer. Only stars are closer. Only stars are closer. 'Cause maybe the night that I'm with you is my light, make me know. Only stars are closer. Only stars are closer. Only stars are closer. Tell me what you gon' do to me. Confrontation ain't nothing new to me.
You could bring a bullet, bring a sword, bring a morgue, but you can't bring the truth to me. Fuck you and all your worst petitions. I don't even want your congratulations. I recognize your false confidence and calculated promises. All in your conversations. I hate people that feel entitled. Look at me crazy 'cause I ain't invite you. Oh, you are boring, you the moral to the story. You endorsing motherfucker, I don't even like you. Corrupt a man's heart with a gift. That's how you find out who you're dealing with. A small percentage who I'm building with. I want the credit if I'm losing or I'm winning. On my mama, that's the realest shit. Love, let's talk about love. Is it anything and everything you offer? I do it for your heart, your heart.
I know it for your heart, your heart. 'Cause maybe the night that I'm with you is my light, make me know. Only stars are closer. Only stars are closer. Only stars are closer. 'Cause maybe the night that I'm with you is my light, make me know. Only stars are closer. Only stars are closer. Skin growing and ego gets you talking. I get involved like a weed balm. Got no end game, got no weeds off. Got to sleep down. It's the way that you're making me feel like nobody ever loved me. Like you do, you do. You got me feeling like you're trying to get away from me. If you do, I won't move. I ain't as cry for no reason. I ain't as pray for no reason. I give thanks for the day, for the hour, for another way, another life, baby.
I did it all 'cause I feel good. I wouldn't do it at all if you feel bad. Better live your life. We're running out of time. Let's talk about love. Is it anything and everything you offer? I do it for your heart, your heart. I know it for your heart, your heart. 'Cause maybe the night that I'm with you is my light, make me know. Only stars are closer. Only stars are closer. Only stars are closer. 'Cause maybe the night that I'm with you is my light, make me know. Only stars are closer. Only stars are closer. Only stars are closer. As he's done, meet me on the floor tonight. Show me how to move like the water in between the dancing lights. Be mine, be mine.
As he's done, I wanna be nowhere but here with you now. I wanna be one in this space. I wanna be tangled and wrapped in your cloud. I wanna be close to your face. But tomorrow can wait. Freezing time in this place 'til the sun is awake. Like a magnet on me, I don't care what they say. We can do it our way. And if love's just a game, come and play. As he's done, meet me on the floor tonight. Show me how to move like the water in between the dancing lights. Be mine, be mine. As he's done.
Welcome back all. After the coffee break, I really hope you enjoyed Tjerrie's presentation about Future Ready. I'm actually pretty proud of what we're doing here, and it also gives a lot of energy to the company. I'm so proud that I even asked Tjerrie, who we internally refer to as Tjerrie GPT, I'm sorry for that, Tjerrie, to build an agentic AI chatbot, or even better, an avatar of David and myself, to take you through the Q&A later on. I thought it was a very good idea, and Tjerrie said that was by all means possible.
But then Robin spoiled the party because he said the market would not yet be ready for that, and he would really still like to have a personal touch. So here I am presenting to you the financial component of our new strategy and our targets, and David and I will do a live Q&A afterwards, or after we have also had the CEOs of our Polish and Romanian business present. Now, let me get straight into the new messages that we are bringing today. We target an OCG of EUR 2.2 billion and a free cash flow of over EUR 1.8 billion for 2028. This would represent a CAGR of 7%-8% for both OCG and free cash flow based on the full year 2025 target, share price of 19 May, and the buyback of EUR 300 million per annum.
Growth for both OCG and free cash flow is primarily driven by Insurance Europe, Netherlands Non-life, and Japan. With a higher share of wallet coming from grower businesses, we will ultimately improve the growth profile of NN Group over time. Moving to two positive messages on life and pensions. Our remittances are sustainable until 2040 without the need for re-risking or without the need for further buyouts. We also do not see a so-called cliff edge anywhere in the runoff pattern. We do not see that before 2040. We also do not see that for the period after 2040. Now, lastly, on solvency, interesting topic, especially in these very turbulent markets. We have actually been able to weather these turbulent markets very, very good lately. A few things come to mind here. We have an under-exposure to the U.S. We have seen the trough in negative real estate revaluations.
We have a very safe mortgage book with an average, bless you, with an average LTV of roughly 50%. We also have a skew towards sovereigns and towards investment-grade debt, so we do have a high ability to avoid defaults. We demonstrated that resilience by a pro forma solvency ratio for April, which came in at 195%. This is at the top end of our comfort range, but still below the sustainably above 200%. However, if our net capital build of 8% would come through, we would expect to get there in due course. If we get there, we would favor small incremental steps on our capital return promise because this would both enable us to grow the capital return promise whilst at the same time also further improve the solvency ratio.
We have ample net cash buildup per annum to make sure we have a very strong cash position and keep ample financial flexibility. Now, before we go to the future, let's have a short look back. We have shown continued OCG growth over the recent years, supported by strong underlying business performance. From 2021 to 2024, we achieved a 7% CAGR on OCG, where we have a target of EUR 1.9 billion for 2025. We achieved a similar CAGR for free cash flow, where we have a target of EUR 1.6 billion for 2025. Strong business performance in Netherlands Non-life and also in our international businesses have really been instrumental in driving this OCG and free cash flow growth, and therefore also instrumental in driving a consistent increase in dividend per share, which reached a CAGR of 11%.
That was obviously also helped by the consistent buyback of currently EUR 300 million that we have in place. We're committed to expand this track record of strong capital return to our shareholders. Now, the basis for doing that is obviously continued solid growth in OCG. Having a diversified and sustainable business that also includes growing businesses to offset the slow and gradual decline of the life and pension business is key to that. That's exactly what we've been laying the grounds for over the last couple of years. You can also see that in the chart. On the left-hand side, you see that from 2020, where OCG was EUR 900 million, we expect to grow to EUR 2.2 billion in 2028. We also expect that OCG to become far more diversified, with over 55% coming from the international business, from Non-life, and also from the bank.
Now, we've tried to illustrate the benefits of an improving business mix also for the long-term indication of our OCG trajectory. This all assumes current markets, obviously. If you look at that trajectory, you see a slow but gradual decline of life and pensions. I don't see any cleavage here. I also don't see that after 2040. You can also see that that decline of life and pension can easily be absorbed by growth of the other business units, most notably international, biggest part coming from Europe, but also Japan and Netherlands Non-life. With this continuously improving business mix, as said, we will gear the company towards a higher growth profile in the future.
From a per-share perspective, this OCG growth is further amplified by the continued buyback that we're doing, which on the current promise of EUR 300 million would, as I said, lead to a 7%-8% OCG per share CAGR versus the 2025 target. Now, we obviously also like to see that OCG convert into cash and to convert into free cash flow growth, and we actually do. Our free cash flow target of over EUR 1.8 billion represents a 5% CAGR on an absolute basis and also the 7%-8% CAGR on a per-share basis versus the 2025 target. Similar to OCG growth, also free cash flow growth is primarily driven by the international business and by Non-life. That also means that we have a very natural diversification geographically for this cash. As such, free cash flow will become less dependent on life and pension.
You can see that in the pie charts. It was over 75% in 2020, and it will be below 50% in 2028. Let's move to our business units, where I'll first focus on the biggest contributors for this OCG growth. I'm going to start with our European business, which we expect to grow to EUR 600 million of OCG by 2028. Now, roughly half of VNB is driven by protection products. That's term life, disability, critical illness. These products fulfill a real need. They're not driven by tax incentivations, and we see that that need really expanded due to COVID, and we also see an impact of the geopolitical tensions on that need.
Now, the good thing is that in both CE, but also in Greece, an increasing part of the population actually has the financial capacity to fulfill that need and can therefore address the still prevalent underassurance that we see in those segments. Now, this is really a great basis to continue our growth strategy, but you're not there yet. As David said, you really have to be in front of the customer here, and you need to have a multi-channel distribution network that can ride all trends. Here I'd specifically like to highlight the very strong client agent network, Salesforce, that we have and our highly advanced digital capabilities there that Tjerrie and David already gave some examples on. This really enables us to not only increase the number of leads, it also increases the lead conversion. It also increases retention.
This is all instrumental for further growth of Europe in the protection business. We do not only have protection business. We also have pension services in Europe. This is more a fee-based model, which is really driven by AUM accumulation. We would expect these markets to continue to grow. We would expect inflow to grow. We would expect AUM to grow. Hence, a higher part will also come from this pension business that we have. Now, given that we have had such a high focus on digital in Europe over the last couple of years, we are convinced we can deliver this growth at zero incremental cost by just leveraging the platforms that we currently have. Despite a growing book, we target flat expenses and are therefore outpacing inflationary pressure.
All in all, that would result, as I said, in a EUR 600 million OCG target for Europe, which is a 7% CAGR. We would also expect to see free cash flow grow in line with the growth that we see in OCG, with some scope to improve the conversion ratio versus historic performance. Given the size and the contribution of the European business, I wanted to give a few words on both VNB and income diversification within this business. In a thriving European business, we would expect VNB to grow by 50% towards 2028. In 2024, we actually did a VNB of EUR 254 million. Now, that VNB growth has an impact on OCG beyond the 2028 target. As you can see on the left hand of that slide, the actual VNB only translates for 30% initially in OCG.
That has to do with the long contract boundaries, and it takes eight years before it's fully reflected into OCG. Out of that VNB growth, also in the years after 2028, there will be a natural source of OCG growth to come. We also wanted to give some insights into which countries are the more dominant growth countries within Europe, but we're convinced that all of our markets will continue to have strong growth in Europe, except for the Belgian market, where we would, given the mature nature of that market, expect modest growth. I'd also like to specifically mention Paweł and Kuldeep, who are CEOs of both Poland and Romania, who will, after my presentation, give some more insights on how they would expect Poland and respectively Romania to deliver on this growth. Moving to Non-life.
David already highlighted the very strong foundation we have there, both in terms of composition, 2/3 P&C, large chunkers in life, as well as in distribution. We have the number one broker network in the Netherlands, plus we distribute using four out of the five largest banks. On this very strong foundation, rapid progress in the digital transformation really adds another layer of potential to this business as we are implementing AI at scale, as Tjerrie just alluded to. This is very important to us. It enables Non-life, as a primary beneficiary of this Future Ready program, to really lower expenses. The lower expenses NN Non-life will use to selectively gain market share in SME fire and in sickness. It will also use them to absorb the further inflationary pressure, and it can use them to absorb a potential softening of the Non-life market.
Versus 10 years ago, the Non-life market in the Netherlands has become far more rational. David alluded to that already a little bit. For the next couple of years, we could expect some softening there, but the Future Ready program we will have in place will really allow us to absorb that. As such, we expect to maintain a very strong combined ratio for the Non-life company between 91%-93%. All in all, this leads to an OCG of EUR 475 million for the Non-life company, which is a 4% CAGR versus 2024. You have to bear in mind that 2024 still had some positive experience variances, so the run rate is actually slightly below that figure for 2024. On free cash flow, we reiterate the message for Non-life that we also give it full year. A conversion rate of over 80% should be possible for this business.
Let's move to Japan, another area of growth. First of all, we're working very hard to finalize the regulatory improvement plan, and we're confident that we can return to 2022 VNB levels by 2028. In March, we launched a new product in a fast-growing long-term savings market. First two months of sales are very positive, and we have multiple product generations in the pipeline. Remember, there are only a limited amount of product approval windows every year, and we use all these for our SME business. Our larger competitors have to use them for their retail business as well, which is the bulk of their business. That means that we have a very solid position in this SME market. To give you an example, in this long-term savings market, where we just launched a new product, we only have two competitors.
We're also expecting a positive impact of the new solvency framework called ICS in Japan on OCG. Under the current solvency framework, which is more like Solvency I, not that economical, higher sales hamper OCG because there is no deferred acquisition cost recognition. That will change under ICS, which will become more economical. That means that under ICS, increasing sales will not have a negative impact on OCG. Free cash flow will continue to be restricted to five-sixths of Japanese GAAP. Japanese GAAP does not recognize deferred acquisition cost, but as David already alluded to, and Marius, our CEO of Japan, can give some more detail in the breakout session, we would expect the recovery of Japan to be gradual. As such, we would also expect both OCG and free cash flow to show a gradual incline, and we would expect remittances to grow in line with OCG.
Now, turning to our stable but certainly not boring Dutch life and pension business. Starting with the basis, NN Group has always had a much stronger focus on DB than on individual life. With recent sales and with a lot of renewals in DB, we actually now have a book that consists for 80% of DB and only 20% of individual life. That's important because the DB part runs off much slower than the individual life part. Higher interest rates also enable us to pick up a bit more yield. Now, the strong starting position with the higher interest rates means that the runoff of that book is slower. It's roughly 2% per annum now. We have a really strong track record in cost reductions at life and pension, and we're convinced that we can make that runoff of 2% also come through in our cost reductions.
We see two pockets of growth for life. First, we would expect a continued growth in DC, helped by the pension reform, adding roughly EUR 45 million of OCG in 2028. Secondly, as already said at the full year results, we would expect some balanced risk-taking towards private credit, adding a couple tens of millions as well to OCG in 2028. All in all, we would expect a modest growth to EUR 1.1 billion for the life and pension company in 2028. Now, for remittances, as you're used to, we expect a stable pattern for NN Life. Note that our remittance capacity is a bit above OCG because we release the SCR at a 100% basis. In reality, probably remittances would roughly be EUR 100 million higher than OCG on an ongoing basis. Few words on pension reform.
We see two main benefits: growing the DC business, which is really the sustainable part of the benefits, and the pension buyouts. Firstly, on DC, the addressable market on DC will continue to grow. We would expect more inflows, and with market developments over time, in general, we would expect a higher assets under management there. This is expected, as I said, to increase the DC contribution to OCG by EUR 45 million in 2028. Longer term, by 2040, we would expect a larger impact of this, of roughly EUR 200 million of OCG. That means that in 2040, roughly 20%-25% of life and pensions OCG will come out of the DC business. This will mainly come out of the decumulation business because on that part, we earn a higher spread.
It is good to bear in mind that in the Dutch pension reform, if you reach retirement age, it is mandatory to buy an annuity, which is a DB-like product. Second benefit are obviously the pension buyouts. We do want to engage if the IRR is clearly double-digit. Every EUR 1 billion of AUM roughly adds EUR 10 million of OCG, assuming that we do a longevity reinsurance on the incoming liabilities. If we would use longevity reinsurance to fund it, that OCG of EUR 10 million would roughly be reduced by 1/3 . Overall, we would think the contribution of buyouts is relatively small from a group perspective, and it is also unpredictable. We have not included that in our life and pension target, nor in our overall group OCG target.
Leon, who's sitting over there, will have a breakout session on life and pension, and obviously can give you more information on both DC, but also on the buyout market in the Netherlands. Over to the last business unit, NN Bank. NN Bank is a digital-only bank. It has a very simple balance sheet. It has savings, it has retail wholesale funding, and at the other hand, we have mortgages on the balance sheet. We also have fee income. We get the fee income by distributing insurance products, by originating and servicing mortgages, and recently we also started a payment function for our clients. The bank probably has a finalization of some large projects to come through, which actually means that we would expect some cost reductions to come through at the bank, also driven by the Future Ready projects that we are actually doing.
We would also expect a modest balance sheet growth for mortgage origination and some expansion of the fee income. However, versus 2024, and even more versus 2023, we would expect continuation of NIM decline given the way where interest rates have been moving, and we would expect that to more than offset the cost savings, the modest balance sheet growth, and the increase of fee income. Hence, we have a slightly lower OCG target for the bank of EUR 110 million versus what we realized in 2024. From a dividend perspective, though, things look better for the bank. By now, we have ended the cycle of countercyclical buffer raises that we've had over the last two years, so we would expect going forward that the bank can be a more stable source of dividend towards the group than it has been in the recent past.
We also transitioned to Basel IV on the 1st of January, which actually provided an uplift in capital for the bank as it reduced the risk-weighted assets for the bank, which would enable the bank to this year pay a strongly elevated dividend. Now, we could get some further RWA optimization from moving to an ARB model, but bear in mind it is a very simple balance sheet, a lot of mortgages, so ultimately the leverage ratio will be curtailing any capital benefit that would come from that because the leverage ratio would ultimately become leading. This concludes the business part of the presentation. Let's move to solvency. Our balance sheet continues to be strong with a resilient solvency ratio, a low leverage ratio, as you can see, a very comfortable cash capital position at the top end of our range, and over EUR 1 billion tiering headroom.
That provides us with ample financial flexibility. In March this year, we also proactively managed our refinancing risk by issuing a successful EUR 1 billion of tier one capital. We also included a tender on the outstanding tier one capital, which is grandfathered until the 1st of January 2026, which had a pickup. Grandfathered until the 1st of January 2026, which had a pickup rate of 75%. Now, this resilient solvency ratio is not something we take for granted. We really actively monitor it, and we also act on it. As you can see on this slide, we've been able to weather very sizable market and regulatory headwinds since 2021 by strong capital generation ahead of our capital return promise and also by management actions. Market headwinds mainly consisted of negative real estate revaluations and widening government spreads.
Regulatory headwinds consisted of multiple U of R reductions, countercyclical buffer at the bank, and various [VOLA] rebalancings. As said, we've been largely able to offset that by our strong net capital build, which is roughly 8% per annum, and we would expect that to be sustainable. We also included various management actions, which was an acceleration into our SAA, where we reduced emerging market exposure a bit, mortgages a bit, equities a bit. We'll transfer that into private credits. Plus, we've been very active in the economically very beneficial longevity reinsurance market. That market actually continues to look very attractive at this point in time. We take a very active approach here. Obviously, the basis for a strong solvency ratio starts with a very solid investment portfolio.
As you can see on the next slide, roughly 80% of our investment portfolio is in fixed income, which should already provide you some comfort. The core of our portfolio consists of mainly Western government bonds rated triple to single A. Our corporate and financial portfolio largely consists of 85% of investment-grade credits, is well diversified by sector, and has a focus on Western Europe. During the great financial crisis, 20% of the corporates in general experienced a one-notch full-header downgrade. In such a severe scenario, we would only expect roughly 4% impact on the solvency ratio, which is by all means manageable. On Dutch mortgages, I think most of you can dream that by now because we've said it many times, but we really think it's a safe asset class. We have an LTV of around 50%.
Customers are generally locked into long, low fixed interest rates, so there is little refinancing risk. The personal bankruptcy laws are such that there is a huge incentive for people to continue to service on their debt. Loan losses during the crisis were very minimal. Bear in mind, the great financial crisis, when we entered it, LTVs were almost double. It was close to 100%, and I was at 50%. In addition, a quarter of the book is guaranteed by the Dutch state, so we really consider this as a safe asset class. Now, we have a high-quality investment book. It does not mean that we do not have any sensitivity, obviously, in solvency.
If you look at the sensitivity that we have on solvency, you really have to take into account that 80% of the book is fixed income, and for fixed income, we have a stock and flow principle on our sensitivities. That stock and flow principle applies to the full 80% of the fixed income portfolio. As you know, we have an ALM policy that cash flow matches our long-duration liabilities with assets. As such, we intend to hold our fixed income assets until maturity to pick up the notional. The negative impact from, for instance, wider government spreads on solvency will over time be fully earned back in OCG. Intermediate spread movements on these assets are economically less relevant for us. Moving towards interest rate sensitivities, by our ALM policy and management actions, we also reduce them.
Changes in interest rates do impact the benefits we incorporate from the U of R extrapolation. Lower interest rates increase the U of R benefit and lower the OCG and vice versa. Also on interest rate, there is a stock and flow dynamic, and we're not economically that worried about that, unless obviously we would go back to a zero-for-long environment. Finally, equity and real estate markets may give some real economic impact, but these tend to be attractive asset classes with a through-the-cycle attractive return, which typically also recovers. That's exactly what we've seen at a real estate portfolio that we hold, where we from mid-year 2022 actually lost 15% of valuation from half-year 2022 until the end of 2023.
That loss was mainly that negative revaluation was largely driven by higher interest rate, which pushed up cap rates and hence led to lower valuations of the real estate book. The underlying fundamentals were always very strong. We were able to largely price through inflation. We still had high occupancy rates. Full year 2024, where we all in all saw an increase of 3%, to us really indicate that we've seen a trough in this real estate book. We have 42% of our exposure in residential, a large part in Dutch residential housing, where there is still the expectations that prices will go up due to the scarcity of housing in the Netherlands. Industrial, 26%, is the second largest contributor of this book. That's mainly logistics. In this sector, we've continued to see a lot of liquidity also when prices were moving down.
We obviously see very strong characteristics of this sector at the moment. To sum up, we would expect positive revaluations on our real estate book. After covering business, balance sheet, let's go to our capital return policy. You're familiar with this sheet. It has featured consistently in presentations over the last couple of years. Our three-pillar capital framework remains unchanged with a comfort zone of between 150%-200% of solvency. If we're sustainably above the 200%, we will improve our sustainable capital return with small incremental steps rather than large lumpy buybacks. This will put us in a position where we can both enhance our progressive capital return and improve our solvency ratio at the same time, given our strong net capital build of roughly 8 percentage points per annum. Now let's have a look at the recent group solvency development.
We navigated well through turbulent markets, and our April solvency remains at the top end of our comfort zone between 150% and 200%. It came in at 195%. If you would filter for the untendered grandfathered RTO1, that will become ineligible on the 1st of January due to the grandfathering, or if we would decide to call that, it would become obviously ineligible at the moment that we would send out the redemption note. In terms of movements, as you can see on the left hand, OCG net of the full share buyback of EUR 300 million plus a pro-rata interim dividend added 1 percentage point. Markets had a negative of - 1%, and other, which included the positive impact of Basel III, Basel IV at the bank actually had a positive impact of 2, which gets you to a pro forma ratio of 195%.
Now we're at the top of the range of our comfort range, but we're not yet at 200%. However, if under normal circumstances our 8% net capital build would come through, we would expect to get there in due course. A natural moment to reassess this and then reassess the capital return promise will be at our full year 2025 results. Lastly, on solvency it is also good to remember that we actively reduce sensitivity. Last year, we changed the valuation method we have for mortgages, which lowered the volatility that we have towards mortgages in terms of solvency. The year before, we actually reduced the steepening risk. We have reduced sensitivities over the last couple of years. We also continue to have strong levers to improve solvency at economically attractive terms.
There are especially longevity transactions that come to mind where in roughly the next three years, we would build approximately EUR 10 billion of liabilities that would be eligible. We have referred a lot to our ability to generate capital ahead of our attractive capital return promise. Obviously, it is also very important to look at it from a cash perspective. This actually looks quite good. Historically, we have been able to add roughly EUR 300 million of cash to have a cash build of roughly EUR 300 million per year. We have no reason to see this deteriorate. That means that in the next four years, we would expect over EUR 8 billion of remittances coming out of our businesses. We would expect to return EUR 5 billion to shareholders without assuming a further step up in the buyback.
Based on our cash flow expectation, that means that we will accrue over EUR 1.5 billion of cash. This really provides us with ample flexibility, cash and financial flexibility for value enhancing, organic or inorganic growth, or if these options would not be available, or if they would not be attractive and would not meet our hurdles, we could increase the return promise to our shareholders. As such, we feel we have ample flexibility to continue our progressive shareholder return policy. Now let me wrap up and conclude. We have delivered long-term growth, and we expect to continue this trajectory. We have an OCG target of EUR 2.2 billion for 2028, and we have a free cash flow target of over EUR 1.8 billion for 2028.
Future growth is largely driven by Insurance Europe, Netherlands Non-life, and Japan, and therefore improves the overall business mix of our company towards a higher growth level. We have a proven resilient balance sheet with a high-quality, diversified investment portfolio that enables us to weather volatile markets, as demonstrated by our group solvency ratio, which came in at 195% and is at the top end of our range in April. This is really important to me. We have a proven ability to generate both cash and capital, and we would expect to do so, which will allow us to consistently deliver both on a progressive shareholder return policy as well as still maintaining the financial flexibility that we would need. With that, I would like to thank you for your attention, and I would like to hand it over to our CEO of Poland, Paweł.
I trust that by the time that he's done, you will be certain you will be as much in love with our Polish business as I am after Paweł has done with his presentation. Thank you.
Thank you. Good morning, everyone. Poland has been quoted today a few times already during the morning presentations. I have the privilege now to really share with you a little bit of the details about our transformation program. Before I dive into the, I would say, key components of that program, let me first introduce shortly myself. I joined the company in the beginning of 2019. Since then, I am leading NN Poland efforts on our transformation program. Before that, I spent a number of years working in the healthcare insurance industry and automotive industries. As you can imagine, these industries are very mature and very competitive.
They have one common, I would say, impairment in both, and that is you either transform or you leave behind. Very quickly after my arrival to NN Poland, we made the assessment of our position with my team, and we clearly discovered that we are not different. We have to change. We have to embrace the new consumer trends and settings. We have to embrace the new technologies, which were creating for us the unique opportunities. We have to capitalize on our very strong position as a starting block. With that said, I would now move to the situation in Poland because I think it is very important to also see that we have started with a very, very solid place. NN Poland is a top market player across all our key business lines we are present. We have very diversified distribution capabilities.
60% of our annual production comes from our tied agents. The remaining 40% is equally split between the bancassurance partnerships and brokers. As you can see, this is a large economy with 39 million people and very progressive GDP growth over the last 30 years. That converted, of course, into purchasing power of our population overall on the national level. Having said that, we also see that Polish people are very trendsetters. They like to use different technologies. They like to use different applications. Therefore, our transformation program originally has been set around digital programs such as creating technology tooling, which is on one hand supporting customers in all our interactions and at the same time supporting our tied agent channel.
In parallel to that program, which was very much boosted around increasing tied agent productivity by approximately 40% during these six years, we have also decided to acquire, when the opportunity came, the MetLife operations in Poland. This acquisition was already executed at the moment when we had a lot of technology already built in-house. Therefore, very quickly, we realized and took a decision that all ex-MetLife or the vast majority of ex-MetLife systems will be decommissioned. The one-off expansion of the MetLife of our network, tied agent network, will be immediately put from the moment of the legal merger of these two entities into our product portfolio and our processes and our technologies. Again, we were able to convert them quickly within the period of two years to our state-of-the-art productivity levels. This integration was also well executed when it comes to the timing integration budget.
More importantly, we also executed on, even far exceeded, original, I would say, base case assumptions when it comes to the level of synergies. When acquiring MetLife, we were assuming EUR 11 million annual savings coming from the cost synergies. The actuals were at the EUR 16 million. That was very much thanks to the things I mentioned just earlier about moving entire MetLife into our technology platforms. Now, how these journeys have converted into the actual numbers, the numbers which you are most likely interested in. First of all, during the last six years, we had increased phenomenally our overall customer base. As you can see, that growth was well above 150%. At the same time, we had a significant step up in our OCG numbers at the country level. 25% of this growth, what you see here, has been actually derived from the ex-MetLife operations.
75% of that growth is our organic efforts. I think that has created a very solid confidence about our ability to transform and lead and stay in the market-leading position going forward. One, I would say, element which integrates both of these KPIs, customer growth and step up in the OCG, also comes from our product strategy. In 2014, we took a very bold decision as a market leader in Poland to actually move away from the unit-linked sales and move to the protection. We were really the first company in Poland who came with the first initial products already in 2015 on the market. Since then, what you see, we have been able to drive our production, our sales with the annual growth rate at 13%. I think that is an astonishing result.
It only shows that we have properly understood and we really nicely hooked our product offer and distribution and platforms with the growing customer needs and future profiles. Recently, we also tapped into the healthcare insurance, which is expanding our position. We are doing this, of course, via the partnerships with the healthcare providers. We believe that this is going to fuel our growth going forward. If I now look at the three years ahead plan, if we could see what are we going to do, I think first of all, Poland is going to immensely benefit from the Future Ready program run by the group. We will be benefiting in extracting the know-how, the competencies, the use cases which Tjerrie was sharing with you on the AI.
We want to plug them into all our three major blocks of activities, so into the lead generation by really doing even more on the digital marketing front and providing even more matching processes between the advisor and the potential customer. We want to also increase the option for customers to not only start online search for the product, but also be able to finish online sales for some of the categories or go for the advisor with the pre-filled information which advisor is receiving. That would clearly increase our conversion rates. We also want to use AI to further increase our self-service platform, which already is in place. Currently, in Poland, we have more than 1 million users of our self-service platform. That creates immense opportunity going forward when it comes to the personalization of the interactions going forward.
These all elements should convert into the significant increase of our key financial parameters, namely in the OCG, but also in the number of customers. We truly believe that we will be crossing 2 million customers number in three years from now and also improving our annual production by EUR 50 million. Before I hand over to my colleague from Romania, Kuldeep, and he will take you through Romania journey, let me share with you a short demo which will be sort of visualizing to you what our end state will look like within the next three years and how we intend to connect the customer, the advisor, but also our self-service platform in a one-only channel journey. Thank you very much for your attention. Let's enjoy the movie. Thank you.
Kate visits LinkedIn on her mobile phone looking for a replacement for her colleague. She finds a post by an NN agent about a new health insurance offer that particularly gets her attention at the moment. Kate clicks a link embedded in the post that redirects her to the agent's page, where she finds out about the product details and checks the agent's reviews. What Kate quickly sees is the segmentation questionnaire. It is simple and convenient to complete. Kate needs to answer three simple questions about her needs: who or what is to be insured, the preferred type of insurance, and what amount she wants to spend on this coverage. After reviewing the initial calculation, Kate decides to buy the insurance policy online. She can easily authorize the transaction on her mobile phone with a code sent via SMS.
She answers a few simple questions from a medical questionnaire in a secure area. She receives a final proposal regarding a health product with a price. Before making a final decision, Catherine wants to make sure that she has chosen the right policy coverage. To do so, she contacts an agent. She chooses an instant hot call, and the agent calls her back within less than a minute. Dorota answers a call from Catherine in her K2 system. In our hot call, the advisor's conversation with Katarzyna is recorded, enabling us to optimize the quality of interactions through integration with cutting-edge machine learning tools. The agent easily finds a lead in K2. The customer clicks and proceeds to calculation in ONE.
In ONE's sales system, agent Dorota can see what policy coverage Catherine has chosen and can modify the offer on the fly, receiving hints from artificial intelligence in the background. After making sure the offer fully meets Catherine's needs, she closes the sale remotely. After purchasing the policy, Catherine logs into the My NN customer area. There she can see a view of the policy details, gain access to the NN Club, where she can receive a gift for the purchase of the policy. In this way, we want to make Kate a loyal customer of the NN brand. In addition, in My NN, Catherine will receive proposals for extending the protection offered by the insurance policy thanks to the next best offer. In My NN, Kate can also make changes to her policy without an agent.
Kate can log into My NN at any time using the icon installed on her phone.
Thank you, Paweł. It's always inspiring to hear about the ambitious plans of Poland. Three and a half years back, I was sitting with David here in this building, and he was talking to me about Romanians, the fact that Romanians are the second biggest cohort in NN Group contributing to the success and growth of the group. Today, I would like to take you on a journey where I want to show, building on our scale in Romania, building on our strong distribution capabilities in Romania, and building on the scaled data, digital, and AI capabilities in Romania, how Romania, NN Romania is starting to contribute to the growth of NN Group. I'm Kuldeep Kaushik, CEO for the businesses in Romania, and I'm really excited to share this journey of Romania with all of you today. Let's dive right in. NN is the largest life insurer in Romania.
Also, we are by far the market leaders in terms of pensions, both in terms of Pillar 2 and Pillar 3. What is even more impressive is the fact that we have grown these businesses organically, policy by policy, customer by customer, to be the market leaders that we are as we stand today. Five years back, based on our scale and the trust that customers in Romania have on NN, we decided to step into the fastest-growing health insurance market as well. I am really proud to say that today we are the fourth largest player in Romania. Based on our strong risk selection and reinsurance strategy, I can confidently say that in terms of profitability, we are by far the market leaders in the health segment that we operate in. Distribution has been a unique strength for us in Romania.
We have the by far most professional workforce that we have in Romania, more than 1,200 active agents that are spread all across the country and obviously taking the NN brand forward. We also have a strong distribution partnership with ING that we renewed in 2024. It gives us a strong diversified distribution capability in Romania. In the last years, we have focused on growing our protection book. We have focused on making sure that our customers are interacting with us with life products, with health products. You see an impressive CAGR growth in that protection portfolio over the last decade of 25%. Our pension portfolio is also in an accumulation phase. As you would see, we have a CAGR of 12% in terms of asset under management growth over the last decade.
As we close 2024, we have more than EUR 11 billion of assets under management in Romania. That contributes to strong fee revenues as we move forward. Now, that scale that we have, the unique position that we have in Romania, combined with our strong distribution and the fee revenues that we have in the pension business, has enabled us to grow our OCG, to triple our OCG between 2019 and 2024. However, as you would see, the market is really underinsured, underpenetrated, and we have a massive opportunity as we move forward to be able to grow in that market. Why do I say the market is underinsured and underpenetrated? We have in Romania more than 19 million inhabitants, out of which only 11% today have a life insurance product.
If you were to look at the insurance penetration, insurance penetration in Romania is around 1%, which is significantly behind the OECD average of 9%. We are also lagging behind the comparative CE countries as well. If you were to look at the underinsurance, underpenetration, underdevelopment in the insurance market against a country which has been growing quite significantly, there has been a lot of economic growth in the last decade. If you were to look at the GDP forecast for the future as well for the next years, everywhere it's clear that Romania will grow strongly in the next five to seven years and will grow ahead of the European average as well. I wanted to share this specific graph on purchasing power, which tells a story about Romania, that tells a story about the affluence, the growing affluence in Romania.
If you were to look at the EU average in terms of purchasing power and you see the journey of Romania from 2015 to 2023, the purchasing power of Romanians has grown from 50% in 2015 to 74% in 2023. What is even more impressive is Bucharest, which has more than 2.5 million inhabitants. The purchasing power of Bucharest is at 164% of the European average at the end of 2023. That number has grown to 190% at the end of 2024. Underdeveloped market, a growing affluent population, which is highly digital and urbanized. That takes us to what we want to showcase as the plan for the future.
If you were to look at the growing affluence in Romania, the underpenetration in terms of insurance, and the unique position that NN has in Romania, we think it offers us a strong opportunity to grow in the next period. That growth is reflected in our numbers that we have shared with you today. We will attract more than 200,000 net new customers. We will have our new business sales, our AP, grow from EUR 125 million to EUR 235 million by 2028. That would result in an OCG growth to EUR 70 million as we stand at the end of 2028. The plans are clearly ambitious. You might ask, what is driving that growth? There are three big engines that are driving that growth as we move forward.
First one is our continuous focus on offering the right value propositions to our customer, growing that protection portfolio that we have been working on, using the strong distribution that we have available in Romania. The second, as I talked about, our pension book. That pension book is in accumulation phase. As I said, we closed 2024 with an asset under management of EUR 11 billion that will grow up to EUR 20 billion by 2028. That will continue to provide us with a strong fee revenue growth in the pension space. Lastly, the Future Ready program that we have been speaking about since morning, that would enable us to deliver efficiencies by the business transformation that is already underway. Let me highlight a few of the key initiatives under the Future Ready program that we've been working on.
First, we talked about our strong consultant tied agent network that we want to continue to invest in. You saw an example that Tjerrie shared this morning of automated call logging. We have been building a lot of capabilities under the Future Ready program to digitally enable the sales network going forward and also empower them with the data and AI capabilities that one of the examples you saw in the morning. We have many more in terms of matching the right lead to the right agent, allocating the orphan portfolio to the right agent as well, and many more that we want to test. The ambition for us is to grow this tied agent network by 20% in terms of absolute number and to grow the productivity of each of these agents by 20% as well by 2028. You heard about the leads generation.
Romania has been very successful in the last year in the ability to attract digital new customers. The acquisition via digital channels has been a massive success story in Romania. We have, at the end of 2024, generated more than 100,000 digital leads. Out of those 100,000 digital leads, we have been able to convert 13% of that business. Let me underscore that number. In any other business, a conversion rate of 5%, I would give my left hand for. To have consistently delivered 13% conversion on more than 100,000 leads is an impressive achievement. What we are starting to do now is to make sure that we can add some of these data and AI capabilities that we talked about in improving the quality of lead, in having hot calls that Paweł spoke about in Poland as well. Ensuring that 13% number stays.
We want to also make sure that the remaining 87% that we are not able to convert, we are starting to attract and gain a lot more from that 87% as well. I talked about the growing affluent population in Romania. Inspired by Poland, we launched a critical illness product last year. That product grew 4x more than the sales that we had in our plans. That product still continues in 2025 to deliver strong growth for us in Romania as well. The success of that product, we have taken it now to Hungary as well. Earlier this year, we have launched that same product in Hungary as well. A good example of share and reuse, the whole concept of Future Ready, where you build once and you are able to deploy in different countries as well. Lastly, Romanians are mobile first.
The country is highly digital. The customers want you to engage them in digital channels. Through our Future Ready program, we have been investing heavily in building our sales and post-sale capabilities to be fully digital. The ambition is that we have more than 80% of our business, of our transactions, of our interactions with our customers go through digital channels. By 2028, we want to be more than 80% digital in terms of interactions with our customers. Lastly, we talked about our customers who are digital native, who are highly urbanized and want to engage with you in digital channels. We have our NN Direct app, which is the self-service app that we have offered to our customers. As we stand at the end of 2024, we have more than 55% of our customers engaging with us on that app. They are active.
They come and they do transactions on that app as we stand today. Our ambition is to by 2028 have more than 80% of our customers interact with us on that mobile app. That app in the financial services industry is right behind Revolut. In terms of the appreciation for that app, we are competing with the best of the best. Let me summarize with where I started. If we were to deliver this ambitious plan, Romania will more than double its OCG by 2028. As I said, we'll start to contribute to the growth of NN Group, to the growth of NN Europe. Thank you.
Thank you, Annemiek. Thank you, Paweł. Thank you, Kuldeep. While the stage is being set for Q&A, there's a minor issue that we've given you 15 minutes more information than planned. We can do two things: shorten the Q&A or shorten the lunch. I'm guessing the preference will be to shorten the lunch. It's not going to be a cheese sandwich. You're going to have a better than typical Dutch lunch. I do think 45 minutes will suffice to do the lunch. David, Annemiek, if I can invite you to the stage, then we can start the plenary Q&A. I already see hands going up. I would like to ask you.
It's good that you're not shortening the Q&A.
Mention your name, mention your company, and try to limit yourself to two questions, which is always a challenge. Where shall we start? Farooq, maybe.
Hi. Thanks very much for your excellent presentations. I just want to start on an IFRS topic. I mean, a lot of the things that you're talking about are very IFRS positive. So you're talking about new business growth. You're not talking about risking. Can you just maybe steer us about what are the differences in the development of OCG versus IFRS going forward? I mean, for example, is there a VNB timing issue? What about Japan ICS? Does that change how we should think about converting to IFRS? That's kind of topic area number one. The second one is kind of around inorganic topics. What your latest thoughts are on disposals versus areas that you think you really could strengthen, because it feels like you've got a reusable kind of technology in Europe.
It feels like there are more markets you could probably do that in. Lastly, just exactly what do you mean by incremental increases in capital returns? Does that mean step up and then plateau or small step ups? Just give us an idea of how we should model it. Thank you. Sorry, that was three questions.
It's very good that you did not shorten the Q&A, but you shortened the lunch, Robin. To start with IFRS, first of all, to take a step back, we really steer the company on Solvency II, right? Because that drives the solvency creation. It also drives the way that we're regulated, and ultimately that will drive capital returns as well. We do look at CSM. And obviously, over the last couple of years, we have already seen a positive CSM development. We do have an organic CSM grow. It was a bit subdued. Last year, it was 0.5% from the top of my head. That's also because we have the improvement order in Japan, and we obviously do not have a lot of new business there.
As Japan picks up, we think that structurally we'll have a positively growing CSM basis. That is more dependent on Japan actually re-kickstarting sales rather than the change in ICS that we would get there. Yeah, in general, as I've said, we're really focused on OCG conversion into free cash flow to steer the company. Disposals?
Yeah, disposals. Yeah, Farooq. I think in general, if you look at these plans, these are all based on, let's say, organic growth of the business. We haven't assumed any disposals or any M&A in any of these targets. I don't know if you want to put buyouts under M&A, but also buyouts are not, the potential buyouts are not included in these numbers. That doesn't mean that we're not going to be actively looking. I think we've shown in the past that we continue to do portfolio analysis. When we see that we screen on financial criteria, new business contribution, 2% or more return on own funds criteria is one part, and the other one is strategic fit. I think in the past, that led to some of the divestments in Bulgaria and Turkey and also NNIP, our asset manager.
We will continue to evaluate Belgium. I think in the past, we did a sale of a backbook. If we see some opportunities there, we will certainly do it. The same on M&A. I think the experience of MedLife, Aegon, Czechoslovakia in the past, also here in this market, has been positive. At the same time, we're very attached to our strong track record here. I mean, it needs to meet strategic criteria, financial criteria, and we do not want to ruin, let's say, the track record that we have. And frankly speaking, there has not been a lot out there in the market either in terms of M&A deals. All this is based on organic growth, but you can count on us that whether it is portfolio evaluations or M&A, if we see an opportunity, we will certainly act on it.
On incremental increases in capital return, we really like to have a sustainable progressive capital return. With sustainable, we mean that if we take a step, we really want that to be sustainable. I think the best way to think about it is when we had the full year 2023 results. Obviously, there was a lot happening. We just did the unit-linked settlement. We improved the quality of the underlying capital, et cetera. There we raised it with EUR 50 million, which seems like a right step.
That does not include recurring steps as well. I would add to that. Cor.
Thanks for the presentation, Cor Kluis, ABN Amro , ODDO. A couple of questions. First of all, a question on the pension buyouts. We have seen quite a lot of parties doing pension buyouts. I think NN has been quite cautious in that sense, and you have not included any pension buyouts in your budget, which could be a plus. Could you give your view on the pension buyouts? Because you are probably not satisfied with the returns that you see in the market, as you would have done more. Also the dynamics. Why is this? What are other parties doing which you are not doing? That is on the pension buyouts. On M&A, apart from possible divestments, more focusing on acquisitions.
You've done, of course, nice acquisitions in the past, especially in Eastern Europe. Where do you see opportunities? Is something pending? A lot of these cases in general, no names obviously, but where are you looking for? Is it really Eastern Europe, or do we have to look to other countries and regions as well? Last question, maybe on solvency here. You gave the solvency at the end of April. I think stock markets did a little bit better since then. Yeah, could you give a little bit? It's already one month later. Update on what it is today.
Probably it's higher.
How much more detail do you want to have, Cor?
You thought Farooq takes three, I'll take three.
Yeah, yeah, you set the stage. The solvency at 12:00.
Let's start with buyouts. Yeah, if you take a step back, so we've always said we believe that the whole pension reform will be back and loaded. And actually, that's happening. Last year, there were EUR 2 billion of deals done. Depending a bit on how you count, we're at EUR 4 billion today, let's say year to date in this year. There's another EUR 3-EUR 3.5 billion of RFPs out there. Doesn't always materialize, but it's possible that a few of those will also land this year. The combination of that might take you to probably at least, let's say, EUR 6 billion in total plus a few, so maybe EUR 7-EUR 8 billion in total in two years. If you go back then to the EUR 25 billion, that indeed means it's back and loaded.
You would expect much more transactions than in 2026 and 2027 if you want to get to the EUR 25 billion, meaning EUR 8 billion-EUR 9 billion a year maybe could be possible. You're right. We have been very disciplined in our pricing. We want a double-digit return or a double-digit IRR. Now, there's a bit of debate on what exactly how you calculate this. Do you take 100% solvency into account or your commercial capital, spread assumption? It's not a very comparable metric always, to be honest. We haven't seen very attractive deals lately that are well above the double digit in the way that we calculate it. Therefore, we've been careful in engaging it. We don't think it actually adds then a lot of value given all the other deployments of capital that we see. It could change, right?
I mean, if pricing improves, and which we do expect over time if volumes become bigger, it's very possible we'll do a bit more buyouts, assuming that they're going to be attractive. This is also why we haven't put it into the plans. It's lumpy, right? You can have a bigger buyout or a smaller buyout. That is why we haven't assumed it in the plan. What is a very important message is that also we don't need buyouts to sustain the cash remittance out of NN Life towards 2040. It would be an add-on. Maybe if I can add one more point to that, I think, I mean, there's a lot of questions always about buyouts, which I think is understandable. I would also point out to, let's say, the immediate annuity development market.
I mean, if you look at last year, there was about EUR 2 billion in the market. We did EUR 700 million. The way the Dutch system, of course, works is that when you retire now in DC, you mandatorily have to buy an Immediate Annuity. This market by itself has been growing around 5%. We do expect a real step up of that market, potentially growing to around 10%. The reason is that you have your normal growth. People are in DC for a longer period, means that they accumulate more value and the account value is bigger. When pension funds go through the transition and they go to the DB, they choose a DC version, a shopping right is there. That means today, a lot of the capital will remain within the pension fund. When people retire, pension will provide an annuity.
This will change. That means that people can start shopping. That means it's also a market for us. Now, this market, you could call it a collection of small buyouts. Last year, we did EUR 700 million. The IRRs have been a lot more attractive in that market than in the large buyouts. That is certainly part of our plan to continue to be active in that space as well, especially since that EUR 2 billion, we do significantly expect to grow that during the pension reform. Yeah, in terms of acquisitions, yeah, I mean, we've never ruled it out. I think, I mean, Farooq was asking the same thing. I am convinced we're building up some real capabilities that are scalable, and you've seen the examples. There are capabilities scalable, and that is potentially a bit new.
We had capabilities already while we felt we do a good integration. You might, in the AI space, build capabilities that are scalable, and that could play a role in actually extracting more synergies than another owner of this company could. Likely, it's usually still in the markets that we operate. Most of the markets, Romania, notable exception, or the Netherlands, we're not in a position where we cannot grow from antitrust reasons. Often, these deals work better when you have inbound synergies than having to go into a new market where you have limited synergies. We're not fully ruling that out. Like I said, it's a bit theoretical because the M&A market is relatively slow. There seems to be a lot more people willing to buy than to sell lately. That could change over time.
Solvency on what is it today, the 27th of May? We do not have that figure, Cor. We have just been giving 2028 targets, and we have given the April solvency, and now to give an intermediate, we do not. I think equity markets obviously have improved a bit versus the 30th of April. We have this mid-cap portfolio, so we would really have to look at how all those companies did.
Who else? Maybe you are at the front, Ian.
Hi, thank you for the presentation. Hi, Ian Lapie from Gabelli Funds. Thanks for the presentation and congrats on the great financial results last couple of years. To first, what would be the major risks to the stable remittances from NN Life? Would it be lower interest rates, or what else do you track that could change that? Second question, on the AI, you mentioned mistakes along the way. Maybe you could give examples and what is the risk? What type of risks do you monitor if there are further mistakes, bad information provided to potential customers, et cetera? Thanks.
Tom, if you can give the mic to David, then we can continue with David afterwards.
Yeah, I think to start with your question on life and on the stable remittances, obviously, if you look at it from a solvency perspective, interest rates are fairly hedged, right? We have limited sensitivities there. We've also explained the Stock and Flow principle, which is on 80% of the book. I think the thing that could really hamper their solvency is if there would be actual real losses. So there would be real defaults or real losses come through, which currently we would not expect. If you would look at it from an OCG perspective towards the future, then going back to a zero interest rate environment for a very long period would obviously be negative. Just based on the sensitivities that we have out there, you would really have to think about severe default scenarios who could trigger something.
Yeah, I think on AI, yeah, you have a minute. I think there's a lot of learnings here that we had. When we had a center of excellence and Tjerrie was running around, we hadn't educated the busin ess unit.
These guys would come up with all sorts of use cases and the business unit would say, "Great, but I'm really busy right now with something else." I think one part was that we did not properly educate ourselves on this, and we led it way too much on a center of expertise. Data governance is not a cool topic, but it is a real problem. If your data definitions are not aligned and you have not clearly defined, how do you expect to scale things across AI applications if your data governance is not in order? I think another one is, and I mentioned it already briefly, you can come up with all these cool cases, but if they are not scalable, then it is very cool to show in a setting like this. If you cannot scale it across units, then you will not be able to monetize on it.
I think we're very disciplined now on monetization. We come with a case of what is the business case, and it cannot be cost avoidance or something. If we do this, we avoid that or something. It has to be real monetization. I think that was also a learning. I think what is going forward remains very important is the ethical framework that we have. I mean, we have a framework, a criteria that we ask each other. There's a couple of questions that we debate with each other. One is, of course, can we technically do it? You also get a question, is this legally allowed? There's a lot of regulation, obviously, in Europe. The third one is, do we want this? I mean, can we explain this if this goes into the newspaper? How would customers feel about it?
Which is a much more difficult threshold usually than is it legally allowed? We monitor. We monitor anti-risk selection. Is there things happening that we consider unwanted? This ethical framework is something that remains very important. Of course, we still have humans at the end of the day taking a lot of these decisions and backtesting some of the AI developments that we have. That gives us at least a good foundation to have the internal debate on, even if you technically can, and even if it is legally allowed, you might not want to do it. I think those are the debates we have internally.
Did it answer your questions, Ian? Yeah. Thank you. David, over to you.
Thank you. David Barma from Bank of America. Firstly, on remittances and cash conversion. The plan is mostly based on remittances growing in line with OCG, especially for Dutch and Non-life and Europe. What sort of conditions or environment do you need to see to grow that faster? The conversion ratio is still quite a bit below the group average for those two units. That is my first question. Secondly, on the OCG target for Non-life, that is quite an impressive growth from the 2024 kind of underlying OCG base, 5%-6% in a market that is now pretty mature and with pricing slowing down and a combined ratio target that is stable. Can you talk a little bit about the building blocks to get to the OCG target in Non-life? I will be a good guest. I will stick to two questions.
You can pass it on to who.
You want to have lunch, yeah?
If you're starting with Non-life, Non-life actually was close to an 80% conversion ratio free cash flow into OCG already last year. We would expect that it can be and it can sustain an 80% or slightly higher conversion ratio. It has a solid solvency base, and we would expect it's a short-tail business. We do not expect to need a lot more capital there than we already have to sustain the growth. There we're very comfortable. On Europe, obviously, last year, the reported free cash flow into OCG was a bit lower, but then we also had a capital injection that we did into Belgium, and Belgium did not pay a dividend. If you would correct for that, we would be at a higher base. I think going forward on Europe, probably the conversion would roughly be around 60%-70%.
It's always a bit, there are many countries, and there can always be some regulations that either are related to tax or that are related to dividend restrictions. We've seen them in the past, but by now, we've diversified the European business so much, and all of the components are actually growing. We are really confident that within this diversified base, we can still make the remittances that we are projecting. As such, we would expect remittances of Europe to really grow in line with our OCG, with some potential for further conversion there.
Yeah, I think on Non-life, yeah, there's a couple of building blocks. Obviously, we always said we feel that this business should be able to grow GDP plus. A couple of elements I think are important here. I mentioned products. We do see more opportunities to grow in thickness products and in fire, which typically also have attractive combined ratios. You're right that we don't aim to further go down in combined ratio. I think there's also a trade-off between value for customer and a good return for us. We do see in some of those products further opportunity to grow. We see opportunities in, let's say, the channels, the banks that we work with, and the direct business OHRA, which both can write attractive business, and we see growth there.
We spoke about expenses. Quite a few of the items you've seen in terms of expense discipline, digitalization, means that we should be able to really control the expenses also in the Non-life business very well too. That should also add to some of the offsetting trends where you see clearly wage inflation, claims repair cost going up. There is just the general premium increases we see overall. The business grows, GDP grows, and so the overall book will also therefore grow. Obviously, if you keep your combined ratio in check, that will also increase your OCG. I think those are important elements that should enable us to further grow the Non-life business to the EUR 475 million target that we gave.
It is good to see that you think it is impressive. Thank you. Nasib?
Thanks. Nasib on from UBS. Firstly, Annemiek, on your slide 20, the cash position that you are building up to EUR 1.5 billion, I believe that is above the top end of the range of EUR 1.5 billion. If I kind of bring it down to the midpoint, you're at more than EUR 2 billion. Why do you remit so much when you get better returns in the business units? Why kind of get to that EUR 1.5 billion-EUR 2 billion? You put deleveraging as a potential use, but I thought you're happy with leverage. Why is that an item as a potential use for the EUR 1.5 billion? Second question on Japan. I mean, you mentioned that there's certain windows in which others can actually put in product approvals.
There are two competitors at the moment. Do you expect there to be more competition in the long-term savings product? What's the IRR on this business? You say double digits, but I think previously you've said 14% on some products is kind of towards the top end of the teens. Thanks.
Can you pass the mic on to Andrew, please? Thank you.
Shall we first do these questions before we also take on board Andrew's questions? The cash buildup, indeed, if the cash buildup would come through, we would obviously be above the top of the range. What we currently see, we leave sufficient cash and capital in the businesses to sustain their growth. They can actually deliver it with the current capital that they have and with the remittances that we have projected. If they would come to us and say, "We have a strong organic or inorganic opportunity that has an attractive return, we'll support that." There is no issue there. What we have now in the plans is all based on what the current capital base is and on the remittances that we have. That already offers significant growth.
The leveraging, we'll look at it opportunistically if we would do so, but maybe there will be some organic or inorganic growth opportunities that would actually have a higher return, and then we'll look at that. It's just really nice to have that cash flexibility.
Yeah, on Japan, yeah, if you take a step back, so the overall, we used to, or we mainly competed here with Nissei, Nippon, and Daido, and ourselves were the top three in the market. You had, let's say, MetLife, AXA, Sony, Meiji, Yasuda. Specifically in the long-term corporate life, it's primarily Sony and also AXA. We now entered also into this market. Now, our expectation is that over time, we will see more entrants, but especially when you talk about the larger players like Nippon or Daido, they need to make trade-offs on, "Am I going to introduce a product here, or is it going to be a retail version?" They have to make more trade-offs because, like I said, the windows to introduce a product and get pre-approved by the regulator is limited.
Realistically, we're going to see more entrants, and probably we will introduce more products as well in the long-term space because it's not just one product over time. I think we will introduce more products, but it makes us positive on the growth of that market. In terms of returns, yes, you're right. I think in terms of scale, it would come down a bit, but for this, we would also expect returns around 14% is reasonable to assume also in this long-term COLI space. That is on the assumption that this business starts growing again, which we're convinced that we can do because the market growth is there, and we've really shown that we're very capable in managing distribution channels. We don't have a tight agent channel there. 65% actually comes through what we call independent agents, insurance agents, tax advisors.
We have around 25% with banks and then Sumitomo, also another 10%. The business has built a very strong track record in managing these independent agents and these channels because it is not that easy to sell protection products in the SME space. It requires a lot of training, a lot of education, and this has actually become our core competence. Now that this market growth is there, yeah, we should be able to build up volume again, including the IRR of 14%. It keeps Japan interesting because it is EUR 100 million that we can deploy against 14%, which, as great as Romania is, you cannot deploy EUR 100 million of capital there.
Andrew Baker from Goldman Sachs, thanks. Thank you for taking my question. The first one, again on Japan, the International Capital Standard next year, what benefit are you expecting to the stock of capital, both for the local entity and the group from that? Secondly, on the Future Ready program, the payoff of the EUR 200 million that you mentioned, are you able to give a sense of which divisions benefit from that the most? Maybe on the free cash flow. Can you help me with the bridge there a little bit? You have EUR 1.6 billion this year as EUR 50 million or so of investment in the Future Ready program, so EUR 1.65 billion.
You then get the EUR 200 million benefit coming through. Does not that get us already to EUR 1.85 billion before you talk about growth? I appreciate offsetting some of the runoff. Just any sort of bridge there would be helpful. Thank you.
What do you mean with the EUR 200 million of benefits on free cash flow?
Future Ready benefits.
Oh, sorry, exactly.
To start off with Japan, obviously, ICS will have an impact on OCG. It will be beneficial. We will not see an uptick in sales hamper OCG anymore. We would expect it to be beneficial for the solvency ratio in Japan. Under group, it could potentially have a small negative impact on the solvency ratio, but we'll have some management actions to address that. That related to ICS. I think on Future Ready, on the benefits, you want to take it?
Yeah, sure.
Can I take it?
Yeah. Ultimately, all these units should benefit. I think you've seen some of the examples when you talk about it, more than chatbots or some of the other elements. All of the units benefit in the short term. You should expect certainly the Non-life company, as we mentioned, but also the bank. Europe, clearly, not just on the expense side, but also on the growth side. As you've seen, some of the examples are units that benefit. Some of the Future Ready benefits are also in keeping the cost per policy on the life company flat, right? If you can really increase your customer servicing, you get more digital interactions. It saves you also on cost. Business runs off at 2%, but you still need to run down the cost therefore also in line. It also contributes to keeping the cost per policy flat. Yeah.
If you then look at the EUR 200 million of benefits that we see, roughly half of that would actually come through OCG. That would be primarily in the Non-life and in the bank area because those are the shorter duration products. We would see roughly EUR 20 million in VNB, which obviously will come through over time. The remainder will obviously gradually feed through in solvency. That is primarily into the European business and into the life business.
Now, all the targets that we set today, so the EUR 2.2 billion of OCG and the over EUR 1.8 billion already incorporate all of the benefits of Future Ready. From a cash perspective, obviously, as David said, it was EUR 450 million of investments. We already started this program last year, so EUR 70 million of the investments we already did. We already took them last year. That means that we still have roughly EUR 350 million to go. This will not impact the remittances that we get out of our business units. It would probably impact group by roughly EUR 50 million for 2025 and 2026 in terms of cash impact. All the targets that we have are already incorporated by Future Ready.
Yeah, so do not deduct it from the EUR 1.6 billion.
Got it. Thank you.
You added the EUR 50 million to the EUR 1.6 billion, which, it is an investment, right?
Yeah.
Michael, it's going naturally already, so. Fantastic.
Yeah, I jumped. Michael Huttner from Berenberg, had two questions. The one is on slide D4, and the other one is a kind of make us dream kind of question. On D4, there's a bump.
Slide D4 again. Can someone put it on?
Inside D4 in the life, there's a little bump, speed bump maybe. It actually means you're flat. You're not going down. Can you maybe talk about the profile of the life OCG of the next two, three years, I guess? Because it looks nice. It looks nice and we'd have thought. The other question, which is a make us dream question, and you might say, "Oh, why should I make you dream?" Historically, you've actually grown much, much faster than the numbers you're showing here, right?
I don't have all the numbers, but DPS at 11%, and you've got total OCG growth now of 5%, but previously it was 7%. Clearly, there's a bigger engine within NN than we've seen in the past. I'm just wondering which bit of it is the conservative bit. I guess some of it is you're not putting all the benefits of the annuity growth. You talked about immediate annuities, but my guess is you've put none of that in there. I just wondered whether you could kind of flesh it out a little bit.
Let me help you a bit here. We did actually have higher growth if you would look at it for the last couple of years. I think there we had a 7% absolute growth also on CAGR, absolute CAGR 7% on OCG, whereas now it would be 5%, but then on a per share basis, it would be the 7%-8%. You do have to bear in mind that over the last four years, we also included some acquisitions. We had the MetLife acquisition, and we also had some years where we saw rapid interest rate increases, which was beneficial for the European business. It was also beneficial for the Non-life business. We were helped there a little bit by inorganic growth, but also by this year where we had rapid increase of interest rates.
It may happen again, but we're not banking on that one to give you something. Now, on the OCG for life, I think the bump is a little bit an impact of that we actually forecast that by a business plan and then afterwards we run with a kind of a steady growth rate. We do foresee for life the OCG to grow by roughly EUR 50 million because we now have a target out there of EUR 1.1 billion. That is a combination of a little bit of a tailwind because last year we had some negative experience variances, which we would not necessarily expect to continue in the future, slightly offset by lower value of new business, new business contribution because that was still a bit elevated last year.
Combined with the DC buildup, combined with a bit of the re-risking that we still have and that we are expecting for 2028, that would then get you a little bit to that bump. After that period, we would expect the status quo investment portfolio. You also would not have that tailwind from a bit of re-risking.
Hi, good afternoon, Thomas Bateman from Mediobanca. I just want to understand a little bit on the longevity deals that you could do because I think you have alluded to the fact that that could fund buyouts in the future. What is the kind of internal calculation you are making in terms of the cost and what you might be giving up in terms of OCG there? The second question is just on the Non-life side of things. You alluded to the market softening there. What is giving you that kind of outlook? I still see pricing going up in motor at the moment. Maybe if you can just tag on to that as well, the regulators investigated margins for lower customers, how do you think that might impact the market and where does NN stand there?
To start off with longevity deals, we obviously look at the trade-off between capital and expenses for OCG. What will be the cost for OCG? We look at the cost of capital of these transactions. Quite frankly, the cost of capital has been extremely low for these transactions. That is really driven by a huge demand from U.S., but also European insurers. Last few rounds really also with U.S. insurers for longevity risk as they have so much mortality risk. The Dutch very long liabilities on a very neatly administrated coherent book just are very appealing to them.
There is also a bit of a notion potentially of a different perspective on longevity and the lingering impact of COVID on that on both sides of the Atlantic, which if you would then use that and recalculate that, that would make those deals from our perspective and on our assumptions even more attractive. We really look at it on the metric of cost of capital, and we clearly look at what is the capital benefit, but also doesn't it come too much as an OCG expense? Non-life?
Yeah. Yeah, we didn't mean to imply that the marketing is softening. I think some of the things we're doing are preparing ourselves also in case the market is softening. That's why expense discipline and risk selection will play an important role because we cannot assume that things always stay the way they are. If you look at what's happening today, then clearly motor is under pressure. We've been increasing premiums significantly ourselves. Most of the market is, not everybody. The interesting thing about motor is any trouble in the P&C market always starts in motor and certainly in retail motor. We've been quite vigilant in increasing premiums in the majority. The market is doing that, but not everybody. That's a fact of life.
That's why we don't get very bothered by that if we would lose some market share on motor if it's not attractive. I think today it's still running well, but some of the other products, we do less premium increases than on motor. If you look at it from a regulatory point of view, we're comfortable with the 91%-93%. I think it's a good balance between, let's say, good return that we make, but also a good return for customers. Otherwise, we wouldn't also be giving this guidance for over a longer period that we believe that this is the right combined ratio to steer the company on.
Benoît Pétrarque from Kepler Cheuvreux. Two questions on my side. The first one is on the 200% solvency ratio at group level. Do you want to be relatively close to that number, I guess above that number, but relatively stay close to that number over the period, or would you not mind to be running at a higher level in line with some peers? Question number two is also on the capital side. Could you update us on the local ratios in Europe of the kind of units where there's a capital constraint? Also on the Dutch life ratio, 185%-190%, is that going to be a decent range also for the coming years? Thank you.
On the group capital ratio, we obviously want it to be sustainably above 200%. As I said in my presentation, if we look at the net capital build that we have of roughly 8% per year, and you combine that with small incremental steps of an increase in the capital return policy, that would really allow us to both increase that return policy, but also have a bit of an increase in the solvency ratio. That is where our thinking is on the group solvency ratio. Local ratios of Europe, I don't think we've disclosed them, but we at the moment do not see any constraints in any of the European countries to actually get the remittances out that we have in line with our plan. For NN Life, it was at 187% at full year results, still also very much at a strong level. Yeah, nothing more to add to that.
Jason.
I think that side is getting anxious.
Jason Kalamboussis, ING. The first question is on the Dutch life in 2040. What are the assumptions for the accumulation on the margins you are assuming versus the ones that you can see now? At what level do you think there will be? The idea is that normally they should be increasing quite a lot along the way. Instead, to know what you have baked in. The second is on the longevity. You have EUR 10 billion that you will accumulate in three years. Where do you stand now? Should we take the buyout, the EUR 900 million you have done? Where do we stand there? And is most of it going to be coming out of buyouts?
Please limit it too because we have limited time left and there's still quite a bit of time left.
Just on the same, it's just the market has done so far EUR 6 billion, correct? Is that what we assume? Thanks.
Can you pass the mic on to Hadley for that?
Sorry, I got distracted. Whether the market so far has done EUR 6 billion? Yeah. Last year EUR 2 billion, and now we've done, or the market has done EUR 4 billion. There's still some RFPs out there. It could be that it's a bit more. Yeah.
Yeah, and I think in terms of operating margin on the DC business, we've always said that we would be looking at 15-20 bps there. Obviously, the accumulation part is at a much lower level, and the decumulation part is at a much higher level. In the prospects we gave you, where we would expect that in 2040, contribution of DC towards the OCG would be roughly EUR 200 million, we've taken into account that then the composition will have changed and there will also be a bit more decumulation in that figure. You also asked a question on longevity, I think. On longevity, we said we would build up roughly EUR 10 billion in three years. That first year of the three years is already this year. That's roughly EUR 3 billion per annum that we would build up.
The last transaction that we did was actually on the decumulation business. That is not part of this EUR 10 billion, which is more on the back book.
Hadley?
Hi, thanks very much. Hadley Cohen, Morgan Stanley. A couple of very quick, just checking a couple of things. Firstly, the Japan solvency, you said that the ICS impact may have a negative impact on the group level, but you will look to offset that with management actions. Can you just explain what those management actions might be, please? Secondly, around the EUR 200 million cost savings, apologies if I have missed this, how much of this actually flows through to the bottom line? What is the actual number that is in the EUR 2.2 billion? My second question, is there a, sorry.
I thought that was it.
It is already Japan is one question.
Clarifications. Hopefully it is very quick anyway. I mean, you're equipping all your tied agents now with lots more data and AI and technology and what have you. I'm just wondering if their sort of incentivization or remuneration structure has changed at all on the back of that. Thanks.
Yeah, I think to start off maybe with the first question on Japan, if there would be a negative impact, we would always, and we have done that in the past, and we will continue to look at risk transfers via reinsurance transactions for the Japanese entity. You could also think a bit further. You could also think about including the Japanese business on the internal model that we have at NN Group. There are various ways that we could actually be dealing with that.
On the question of cost savings, out of the EUR 200 million benefits of Future Ready, we said roughly EUR 180 million is expense savings, and then that EUR 20 million is VNB. Out of that EUR 180 million expense savings, roughly half or nearly EUR 100 million of that will actually flow through OCG because that's more the short-tail business. The rest will ultimately be captured in solvency ratio as far as it relates to NN Life and pension and to the European business.
Yeah, and unit cost assumptions of solvency. Remuneration structures, I mean, to be honest, remuneration structures constantly change with tied agents. It is possible that over time, if we manage to get the efficiency much more up, meaning that they are so enabled that they can sell a lot more, that also the commission structures will come down. There will be a trade-off on how effective we will be. But agent retention is also very important. Historically, I mean, agent retention was shocking. 70%-80% of the agents in general in Central Europe are gone after a year. So a lot of effort is now put into keeping them, training them. But yeah, I think the possibilities are a lot. So over time, you could see a change in remuneration structure because of all the support that we're giving as well.
That side, show of hands. Farquhar? Last question? I think you can leave it there.
Farquhar from Autonomous Research. Just two very quick questions then. Just coming back to the Future Ready program and AI more generally, I just wondered if you could give us a sense of how many FTEs work that is doing for you at the moment and where you expect that to go. Secondly, a little bit of a follow-up to Hadley's question about the cost saved in the EUR 200 million. The element that's going into the Solvency II ratio in terms of capitalization, should I think of that as keeping you at pace with the cost assumptions, i.e., not really additive to Solvency II net net? Thanks.
I think on FTEs for AI, when we say it's EUR 180 million of expense savings, and indeed half will go into solvency assumptions, half into OCG, but that doesn't really matter in terms of, yes, there will be FTE impacts, clearly. I mean, there's other savings as well, but part of it will be FTEs. We're really steering, let's say, on the total expense savings. We're also hiring people, but there will be obviously job losses. To be honest, it's nothing new.
I mean, if you go back 10 years, every year we have digitalization has been ongoing. We incorporated Delta Lloyd and VIVAT, and the overall FTE numbers have continued to trend down. We do this. Yeah, when you get to the numbers, this is what you discuss with the union, with the workers' council, you get into a social plan, and that is when you also talk about the actual FTE numbers. It is not something we steer separately on. We really steer on the monetization through euros, not so much on FTEs.
Yeah, and on the question on capitalization of the FR expense, for life, we would really use it to keep the unit costs flat and in line with the runoff of the book. Similar to Europe, although there may be at some point, if you can prove that your assumptions are actually that you can yield the results, then there could be some room. It is mainly used to keep the expenses in line.
Thanks for the question. I mean, we still have four breakout sessions coming later this afternoon where you can ask a lot more questions. I think, David, you want to wrap up for the webcast?
Yeah, certainly, I think for people on the webcast, we still have a very interesting and exciting program that will come offline, unfortunately, for all of you. I want to thank you very much for participating. Obviously, we're proud to present and come out with these targets, the growth that we're saying, the change in the profile of the company, supported by underlying positive trends, but also by, let's say, all the technological advancements that we see. Thank you very much for joining. We will now have a lunch break for all of us, and then we'll be back to go into the deep dive sessions and the breakout sessions. For people on the webcast, I hope we will be on the road a lot, starting travel tonight. Hopefully we can also meet in person in the incoming period. Thank you very much.