Okay, good afternoon, everyone. Thank you for attending this event at the Musée de la Marine. Thank you also to those of you who are connected and listening to us remotely. It has been six years. The last same event was six years ago. The company changed quite a bit since. The meeting was in 2018. We had a smaller event in 2021, very specific to the semiconductor industry. The objective of this meeting is to tell you about the new company, the new Mersen. For those of you who know the company or may not know the company.
[Foreign Language] .
I'm going to present the agenda. Luc is here, ready to start. He will be taking the floor first. He will tell you about the new company, the new Mersen, about the company's strength and winning positioning. We will then talk about midterm growth drivers. He will be speaking with Salvador Lamas, who is in charge of the Solutions for Power Management division, and Olivier Raymond, who is in charge of sales and marketing for Graphite Specialties, the Graphite Specialties business unit. All three, they will present the company's markets. We will then have our CFO. He will tell you about the company's financial profile. Thomas Baumgartner and Luc will then take the floor to close. Of course, there will be a session dedicated to questions, Q&A, and as you probably know, we had a press release this morning.
The slides, as well as the replay of this event, will be available immediately after this event. Luc, without further ado, over to you.
Okay, I hope the microphone works. I hope you can hear me. Good afternoon, everyone. I was able to speak already with some of you. I recognized some experts of the company, people who know us really well. I'm sure that most of what I'm going to share with you is going to make sense and resonate. I hope you were able to take a quick look at some of the products we manufacture that are displayed outside. So the company has taken on a very new dimension. It didn't happen overnight. The company transformed little by little over the past years, up to 2016. We were finding our way, looking to find our way.
We were able to find our way around that period, and this is when markets started to accelerate and the company's growth accelerated. The company has become a more global player with a true global footprint. Our product range has consolidated, not too much, but somewhat consolidated. Today, we operate in all locations where there are industries in all big markets. So by having a truly global footprint, by including markets that are really accelerating, we have become a less cyclical industry, a less cyclical business. I think one year we were down 15%. Never say never. You never know what could happen, but if things go well and normally, we are today in a position to absorb ups and downs. We'll come back to this in greater detail when we take you through our plan. We are today a solid company. We're here to work alongside our customers.
the years, we were a lot on traditional markets when the GDP was good. We knew the GDP would be good for us. We saw new businesses like, we were in semiconductors, SiC, of course, with the emergence of electric vehicles, electrification in general, going from thermal to electric. We never had that many sales on thermal so we decided to become very active on some of those markets. Initially, quite modestly, we went quite cautiously. We progressed at a moderate pace so these markets and our traditional markets are doing really well, are very buoyant and all in all, I think I can say that today we are a fairly stable company so I think we've moved on to a new dimension, which is also true when it comes to the financial results of the company.
As you were able to see in the press release, the EV market with SiCs is somewhat lagging behind. It started in February, in March, in Europe, where we saw that the numbers were not that high. Things didn't really improve during the year. In September, especially in October, we felt we had enough information to explain that some markets were a bit slower. So we'll come back to this in greater detail a bit later. So the company took on a very new dimension. Thomas will come back to the numbers, but as you can see, this slide goes back as far as 2017. We then had the COVID-19 years. The COVID-19 somewhat slowed down the business. We then reached an important milestone, the €1 billion milestone, which we exceeded in 2022, up to €1.2 billion in 2023, and we will be above €1.2 billion again this year.
You probably did the math following our guidance, so this is truly unprecedented, and this is going to continue. I think the slide we showed in March showed aerospace and railway, which were really part of the process industry. What's interesting in this slide is that we have dynamic markets, less dynamic markets, but they're all double-digit growth markets, and you can see that the upper parts, the SiC, EVs are really picking up. There is a true momentum, and this is really the business we're focusing on, and meantime, aerospace and rail were very strong, very profitable. We have a roadmap that goes beyond 2035, so the green part will increase. Rail as well. The average is 60, so aerospace is actually above that. Rail slightly below, but still very good levels of growth.
We have the process industry, process industries, which contains a lot of different things. You were probably able to see some examples and applications of the process industry. There, once again, substantial growth as well in the vicinity of 11%. That's the company. That is the company's new configuration. As we have told you in the past years, the gray is really what pays for the orange, the development of the orange part of the business. We have become a more profitable business. Thomas will confirm this a bit later. You can see the values here and the incremental increases. We're going to probably go beyond the numbers that are here. 2024 will not be that far from what's here in terms of EBITDA. This shows that we are a company that is with good long-term profitability.
Cash, you can see there was a bit of a trough in 2022. This is where we really needed a new industrial momentum. We had to reinforce our industrial footprint, and we had some buffer stocks as well for the automotive industry, and we had some issues with some suppliers in the U.S. Some component suppliers were a bit deficient, so cash-wise, it wasn't really a good year, but we're back now to a much better situation and should be just as good as in 2024. What happened more recently is that electric vehicles and the new electric vehicle platforms that were due in 2025, which is pretty much now, which were initially going to be equipped with silicon carbide and battery protection equipment as well, so the sales were lower than expected, which had an impact on the two business units.
We've really felt a slowdown since the beginning of the year. There have been a number of announcements since the beginning of the year. There are some restructurings on thermal, some delays on electric vehicles. It's difficult really to pinpoint precisely what is happening. We think we have fairly reliable information because we get orders for graphite and silicon carbide. So when you put all the orders together, obviously, not all customers have the same timeline perspectives. Most of them, however, are expecting two, possibly three years that are a bit slower, which is not what we had initially estimated. We have a hunch that the production of wafer and transistors is going to be postponed approximately three years. We've delivered a lot of products in 2022, a lot as well in 2023. And those companies need to make transistors a long time before. They need to make converters.
Converters have to be then installed upon cars. It takes time. The production of cars takes a good year. So the lead time is very important. So right now, I think we're probably going to have to wait until 2026 for things to really pick up. And then hopefully we can come back to the numbers that we had in our initial plan.
In addition to that, as I said, we have other growth markets that are doing well. The company's investment plan to honor the contracts with the—we had some penalties, late penalties. Our legal department has been very, very active on those. In 2022, we could not really say we're going to invest, and we'll see how things work out. Back then, it was one shot for Wall Street, for example. We had some major investments to make for equipment to actually get things started.
And so I would say 70% of this is ready now. So we're going to have to postpone, but there is not much to be postponed really anymore. So the good news is that since we're ready, we'll be ready until probably 2029. So that's it for the CapEx. I think we're in good shape, at least until 2029. Thomas will discuss this.
We also worked on making some of our sites more profitable. And this is something we accelerated. We're doing our very best to be fully prepared for 2025. 2025 will not be the best year ever. I think we're in good shape, however. We're very ready for 2025. And the rest really remains unchanged.
You will see in the charts and the slides that the electric vehicle, the plan we had for electric vehicles was only for the beginning, for the early years of the electric vehicle journey. But there is a lot more to come, a lot of positive things to be expected. So the company's positioning, I think you know most of this. We're just going to review a few things that you may not be familiar with. First of all, we have our unique set of expertise. If you're connected, you don't see this, but we have displayed some products. We have videos as well that highlight this unique sets of expertise. Then we're in a key position in our customer's value chain. And we're also really looking at developing a more sustainable, a more environmentally friendly planet. On materials, I'm not going to go into the details.
I think you have information on graphite production. It's pretty clear. The material that's displayed outside shows you how graphite is produced. After years of investment, we're number one, 14,000 tons of graphite before the end of the year. Most of those products are consumables. In some of the processes that were showed outside, that can be—so there are multiple applications, formulation, tooling. It's global and unique expertise we have. Just a few other competitors have this. And then our footprint, I think, is industrial footprint is quite unique. The Japanese are very Japanese. They have no plants elsewhere. Our SGL competitors are in India. So I think we're also a bit ahead of them. And we're really taking advantage of countries where we operate. In the US, the cost of energy is half the cost of energy of Europe.
There are some other labor costs, but that is secondary. I think you know some of the applications. Again, you had some examples here in the room. We have a cooler, some applications. On the electric part, we design all our products. Fuses, very few components are made by Mersen. We're doing some plastic, metal transformation as well. We do a lot of research and development, a lot of design, computing. We do more and more testing to have approved certified products. They have to be certified before we can sell them. There are only two companies in the world that can do this. So that's for the fuses. And then that's the part that Salvador manages. Power converters as well, except for the metal parts that we buy. We make everything. We do the design, product design, the bus bars, as well as the cooling.
We have a quite unique offering as well in power electronics. All product lines are today available on all three continents, even more than that, actually. We have some production now in India as well. So we can produce locally there for some customers like Bombardier, BHEL, Alstom, and we really have opportunities in pretty much all countries. So multiple applications that can be used in different areas in speed, variators on cars, on windmills, converters, and more on aircraft. If there is a megawatt of power in an aircraft, which is huge, really, you wouldn't think, and the planes also have bus bars and power conversion systems. R&D continues to play an essential role at Mersen. We do a lot of fundamental research.
We also use the services of external research labs, if necessary, sometimes for very specific customer requirements, or if we need to upgrade a product, or if we need to lower costs or boost the performance. We have 180 experts in the company. We didn't really have this number back in 2018. We did a lot of work to understand our skills and what our skills were. So we take very good care of them. We take good care of everybody. So we have 210 employees in R&D and innovation worldwide. Testing capabilities, electric testing capabilities that are very, I wouldn't say unique, but we have two big test benches in the U.S., one in Lyon. And you were able to see some fuses outside here. That's developed by our R&D team based out in Shanghai, China.
We have more and more people outside of France who are involved with the development of new products. We do digital simulation. A lot of companies do. So we definitely do digital simulation. It has always been something we've done. Today, we're doing it in a more formal way, in a more sophisticated way. We have more computing power. Initially, it was centralized, and you would launch a program, and you would have the result four days later. Today, we have massive computer power with Kheops
That is a project that is funded by DGAC, and the sponsor above that is Airbus. So we're looking at making bus bars for converters under high-pressure environments. Fast Lane is a project which has been in the pipeline for a number of years. So these are contracts between EUR 500 million and [Uncertain/Garbled] .
Then we have some more unusual, more breakthrough technology, starting with Soitec. Soitec was started in 2020. In 2024, the product is pretty much finished. It can now be manufactured industrially, even if the results are not where we would like them to be. But I think if you're familiar with Soitec, we're going to use a very thin layer, a monocrystalline layer, and put it on a cheaper substrate called a polySiC. There are some examples here. We have pictures in the room as well showing this particular application.
It can then be used as a monocrystalline in different types of devices, semiconductor devices, like the semiconductors that are used in vehicles, for example. The key benefit of this application is the cost. Soitec is also claiming greater efficiency, better yields. It conducts heat more than others. There is a real technological edge with a product like this, like this one. There will be on cars. There will be used on vehicles, as I mentioned earlier. I think that this is probably the most complex and the most innovative product we've had in the past years, really starting here from scratch. At the bottom, we have another example of a power storage system, four- or five-megawatt units, 1,500 volts. We have protection systems that are provided on the side and for high-voltage charging systems on highways.
These are as high as 1000 volts of power. In the future, you'll be able to charge a car, even a truck, very rapidly. We were asked to develop different fuses. You can see one picture here. They look very different with potential cooling systems included. Here again, we did a lot of research. We started from scratch. It's not available yet, but these fuses have been tested and they have been approved. Here is a quick overview of our customers. As you know, we don't have one customer representing 50% of our sales. The biggest customer is five. If you look at the top 10, the top 10 cover 25% of the total sales. This includes semiconductors, probably Wolfspeed, and then rail. I'm sure this includes companies like Alstom, solar, probably Longi. In the process industries, I wouldn't even know. Let's probably Saint-Gobain.
Top 20 is 36% of sales. So here you can add companies like Safran, Power Conversion, and companies that operate wind power as well. And then the top 30 is 42%. So this probably includes also some chemical customers. So a fairly balanced customer portfolio. And with a portfolio like this, we limit our risks. Siemens is probably also here, probably in the top 10. Even when we have ups and downs and fluctuations, if you have one big customer, the risk is that you lose a lot of business, which obviously is not the case for Mersen. We are in an industry where you don't see newcomers every day. You see newcomers from time to time, but not every day. We have very good relations with our customers. We have good maturity. Some customers we've been in business with for many, many years.
I think we can say we have a true relationship based on trust. Sometimes there can be some tensions, but all in all, I think we have very good relations. New customers, well, they're not new companies. BMW, Canon, I think it's for printing head. EAE, EAE is a Turkish company that develops data centers. They represent several million EUR every year. Fortescue, EKS, Spanish company working in energy storage. Alpitronic, Alpitronic, they are an Italian company. It's a family-owned business. They launched EV charging units. They went to EUR 3 billion in sales very quickly, so they've ordered a few million fuses and bus bars. You don't really see their name anywhere, but when you see electric cars on the road in France, you can bet that they have components that some of the components on board were made by Alpitronic.
They're a bit like ABB, Okmetic. Okmetic, they also make silicon.
We put it there because it took two years to have the monocrystals . Moving on to a very important slide that you're familiar with, but I really need to stress this. Everyone can do this. Really, we see this because these are special animals, but CAC 40 companies, some of them are less well broken down. For the people to whom we present Mersen, we're not very well known. So we can discover to what extent we're everywhere where we should be. It can be industrial too, with GDP 35% or 34% for semiconductors in SiC. Thanks to them, well, before we had just a few millions, and we went up to 50 or 60 thanks to these large companies. North America, a few words on that straight away. Europe, lots of countries, but 38% of the group's turnover.
Asia, 26%. Solar, not that good right now. We had wonderful years in the past and some months we could be at 30%. We wanted to show you a slide on, I was going to say Trump, no one in the US, just to show you our achievements. Like everyone, things slowed down after 2020 and then picked up, more due to the supply chain. The business was still there, but you can see the acceleration. In this acceleration, Wolfspeed is there. Onsemi is there. OEMs for silicon too. Process industries strong in the US. We captured market shares, 410. Well, to that, we need to add, oh, yeah, there's a slide in that on acquisitions. GMI, Bar-Lo, and CG Thermal, especially GMI, which was the biggest for graphites when we joined that group. $40 million in revenue. Bar Lo, let's say around $15 million.
Let's say you can add. We're going up to 70 if the rest doesn't change, so you can see revenues growing in the U.S. for Mersen. Mersen is almost an American company with standalone teams, and at the end, we'll have questions as to what happens if the country closes down on itself. We have our product lines, and our American colleagues can supply the country if needs to be, and they couldn't buy elsewhere, but we'll see what happens, but it's a very important country for us in a way. Everything is there. It goes quickly. You won't talk about differences once again between the U.S. and Europe. There is nothing on Europe, but one slide on India because of my growth plans. We decided to present it based on the market, but we could have done it based on the country.
In 2027, what are the countries that will generate more revenue for the group? This is one. We started slowly because China was slow. We had to be patient. But we can see there's a small acceleration due to rail and wind. For CapEx, not this year, but the year before, there was a good amount needed to handle rail and especially wind in India. I can't tell you exactly what revenue we'll have in 2029, but about 35. But you need to keep close watch of that country. We can gain tens and tens of millions of euros in revenue in the next 10 years. And they've started to talk about having plants for semiconductors, but they don't want to buy Chinese products for another 100 years. So things are changing. Don't forget CSR.
We also wanted to put this topic on screen without going into the details because we spend - you could spend all day on it. So a responsible partner, okay. That's not a revolution. We have always been mindful of social and environmental conditions. We don't have plants that are very complicated to manage. It's not a refinery. It's much more sound. But there are little issues here and there in terms of employment. In our practices, we have a diversified footprint to offer to our employees, offering the same benefits to our employees, whether they work in China, in India, or in the US. It took some time, but I think we have reached a sort of homogeneity throughout the group, and that's important. So we spoke about the environmental impact of our sites. We work a lot on waste, reducing waste, recycling, water too, a very well-known topic.
So that must be looked at. Developing human capital. So this is really a part of the group's DNA. We don't do just any old thing. We have lots of rules on exports to countries where we are not entitled to sell. So we keep a close eye on all of that. And there you have it. At the end, a few examples of what we can offer to customers in terms of reducing consumption of water, electricity, carbon footprint, to have a product with just the right components for its functions and nothing redundant. So a lot of work is done on our products. We design the cost to be sure we don't spend for nothing. I won't go into the examples.
They're all just technical, but I've never seen an example with Mersen where we don't try to reduce something and helping clients to make their system more efficient. I've never seen anyone selling a product that was not as good as the previous one. When you work on costs, you must review your design and even on functions specific to products in your customer's equipment. We always bring in things that are more lasting, more efficient. As a matter of fact, it's not easy to explain that when it comes to taxonomy because it's complicated to prove. When you give equipment for a furnace, if it lasts for a year and a half, whereas before it lasted for two months, it's hard to quantify that. In addition, most of our markets are end markets that are doing well because there are more semiconductors, of course, more electronics, more digital.
But you also have the grids, electric cars, solar panels, and all of that because society is going in that direction. To give you an example on solar panels, in 2012, 2013, just imagine one gram and how much silicon it needed to produce a watt. It was 12 grams. Today, it's much less. So it's very interesting. Power 12 gigawatts, for example, we can do it with 12 grams. We could put it in the library, 10 square meters, very expensive, subsidized. No, we no longer need subsidies. We're not the only ones responsible for that, but not bad. We played our role. So that's what I had to say for the first part of this presentation.
Over to you, Olivier.
Yes, over to me. Don't forget to press the green button. Thank you, Luc. Hello, everyone.
As Luc mentioned, we are positioned in markets with a bright future at the heart of the major sustainable development challenges for our planet. Our products are not always visible, but they are essential to the major issues of developing renewable energy, energy transmission, developing smart grid networks, and energy storage. In short, we're playing a key role in improving energy efficiency globally and to shape a more sustainable society. To begin with, let us take a look at the renewable energies market. As you know, the energy mix will be largely transformed by 2030. The share of renewable energies will increase. In 2023, it accounted for about 30% of the world's total power generation. By 2025, renewable energies will overtake coal to become the leading source of electricity generation, and in 2030, all renewables will account for about 46% of total global power generation.
If we were now to focus on solar energy, which is the most important renewable energy market for Mersen, some 440 gigawatts of solar panels were installed around the world, which is huge, representing more than half of India's annual electricity consumption, almost all of the total consumption in Brazil. Most of these solar panels were installed in China, about 250 gigawatts in all, as compared to 440 globally, and then we have the United States, Brazil, and Germany. We operate in all those countries. Installations should increase by a factor of three by 2030. We're not trying to keep up with this growth, but this momentum is important for our group. Now, turning to the value chain, for producing solar panels and running solar panels, Mersen is active at two levels of the value chain. First of all, with two different business models.
That is, first of all, in terms of materials, you supply graphite and felt to solar cell manufacturers for their silicon ingot pulling process, a high-temperature process. We're talking about 1600 degrees. For the moment, it is mainly in China, and we deliver from our Chinese sites. On the electrical side, what matters is where the solar panels will be installed. So we provide around the world fuses or surge protection to protect equipment and people around solar panels. We also deliver passive components for energy storage and conversion systems. So for Mersen, there are two important things: where the cells are manufactured and where the solar panels are installed. We're going to see new projects emerge for producing solar panels outside of China. This is linked to the strategy of some countries that want to reduce their exposure to Chinese manufacturers. It is an opportunity for Mersen.
It's hard to quantify at the moment because it remains yet to be demonstrated that these projects are economically viable. Mersen has the advantage of having geographic operations around the world, as Luc said, so we can produce locally to serve the local market. This can accelerate the development of solar energy in the coming years as of 2026. So in conclusion, on renewable energies, we have an unchanged medium to long-term potential, even though we have to manage a less favorable short-term situation, particularly in China. By 2029, all renewable energies should represent more than EUR 200 million compared to around EUR 140 million this year, up by around 4% over the coming years. This should rely on our ability to take advantage of our industrial capabilities around the world to address all of these new projects emerging outside of China.
On our installed base for wind and electricity, we have a significant potential in energy storage and transmission, thanks to our strong relationships with the large OEMs. I will now give the floor to Salvador, who will tell us more about the electric vehicle market.
Thank you, Olivier. I will now talk about an important market for Mersen that lies at the heart of the sustainable development challenges of our planet. The electric vehicle market, as Olivier said, a market that is spoken about in the media with lots of news, good or bad. We'll talk about this in detail to see what this means for us at Mersen. You need to understand that you already operate on these markets in several segments. You have the industrial-heavy industrial vehicles, like for construction works, machines, or Claas or Komatsu as well, agricultural machines as well.
It's mainly buses too, for passenger transport, the class like Bell, or for goods, passenger light vehicles or utility vehicles. We have a third segment, light vehicles, private individuals, let's say, passenger vehicles, characterized by large volumes for a product type, but also a strong growth momentum in recent years and in future as well. Let's focus on light vehicles for private individuals. In recent years, we have seen very good growth, driven by three factors. First of all, China, which represents the largest share of the market today. A strong growth of the market in China, the quickest shift to electricity on that market. Secondly, the second factor, people. People we call early adopters. People switch to electricity. There are some of you in the room or online.
These are people who adapt new things quickly, taking risks on the motorway to see if they will find a charging station or not, but that accelerates the change to electric vehicles, and thirdly, standards in Europe to reduce CO2 emissions, and all of that stimulated the market. Looking at future forecasts, these markets for electric vehicles will continue to grow by about three or four times by 2030, according to specialist firms. The data is for September 2024, so quite recent. Of course, growth will be driven by China. Their market should double in size by 2030. The important thing for us is that Mersen, on the Chinese market, has limited accessibility. For example, we're one of the biggest car manufacturers, BYD, which is vertically integrated for manufacturing components or subsystems.
It doesn't need to buy from subcontractors any components like Mersen, which, of course, restricts access to those markets. We also have the rest of the market, which is growing more strongly by a factor of four by 2030. And they are open to us at Mersen. As I said, we have norms as well. As of 2025, norms such as CAFE, Corporate Average Fuel Economy, imposes a maximum CO2 emission threshold per manufacturer. And they're asking for a reduction over the years. And that should stimulate the switch to electric vehicles. As I said at the beginning of my speech, we hear a lot in the media about the market, with a time lag on the market. So today, we consider that the figures from 2025 to 2027 do not fully reflect the slowdown of this market.
The important thing to remember is that despite a three-year lag in volumes by 2030, the market should triple, at least, confirming the potential of this market for us at Mersen. We spoke about the market situation, and now let us look at Mersen. We operate on the market with two product ranges. First of all, interconnect bars or bus bars systems enabling clients to connect the different cells of the battery to assemble a modular battery pack. And then we have fuses protecting the battery pack. And that can also protect auxiliary functions such as the radio, air conditioning in the car. These fuses also protect people, passengers, all of us, because these fuses are integrated in the systems with our clients, enabling a quick disconnect in the event of a problem with the battery pack or in the event of an accident.
You also have the market for infrastructures, for charging terminals. This is important for us because infrastructures are necessary for this shift. We operate on that market thanks to fuses or high-voltage protection. We put a few examples of clients on screen. Going from the start of the value chain with producers of battery cells, Panasonic, for example, for ACC. We also have subcontractors or tier-one manufacturers in the automotive, with Magna, for example. Moving on, we work with car manufacturers to Rivian, BMW, for example. For charging terminals, as Luc said, we are already a leader in the market. We work with the market leaders. If you charge on a Total station, you will find ABB equipment in the system. And it's Mersen. A highly competitive market, of course.
So to make a difference on the market, we created a complete line of dedicated fuses covering all the voltages on the market. We hear about 400-volt platforms going up to 800 volts, enabling faster charging or a longer range. So we created the entire portfolio thanks to our expertise in power protection. We're number two around the world for electrical protection at Mersen. So we have developed a portfolio produced on automatic production lines. And you know, there's a video showing our production lines in China that are totally automated, recognized by our clients as a highly automated line, even as compared to Chinese competition. Let me come back to fuses. These products are fully qualified thanks to our test laboratories. Luc spoke about our ability to certify our products from design to prototyping, all throughout.
For interconnect bars, bus bars, it's about providing specific solutions because each battery module is different depending on the manufacturer. Even the type of cell or chassis is different. We also integrate smart functions, for example, temperatures for managing the battery. We also provide solutions for bus bars for car converters that are inside cars used to convert between AC and DC. To give an example, to charge your car at night, you connect at home using alternating current. But the same converter in your car allows you to connect the next day to an ultra-fast charging terminal using DC, thanks to our bus bars that make that energy conversion possible. In addition to the video in this room, we tried to show you a picture of a line we received in Saint-Bonnet, close to Lyon, to produce automatic bus bars for ACCs.
This line is brand new and is just starting. It has been qualified with our client ACC, to give an example. It's full of innovation in terms of manufacturing because it will be used to produce a product every 10 seconds. That's so revolutionary as compared to what is done to produce this type of product. Of course, thanks to this expertise, we're already connected to most automotive manufacturers outside of China. For light vehicles in Europe, we'll find French or German manufacturers. In North America, we'll find Ford or General Motors, and we also work with newcomers like Rivian or Lucid in California, for example. We also work with Korean manufacturers, Hyundai or Kia. That's also the case for heavy vehicles, BMW, Iveco, or Scania. We spoke about the three-year lag on the market. Of course, over the short term, we had to adapt and optimize our industrial facilities.
What we did in 2024, for example, we centralized the production of fuses to one site to be more competitive and to react to changes in market growth. We remain flexible and adapting the pace of our investments. That means being able to invest at the right point in time as volumes increase with our clients. And this is something we can do for this type of product. In a nutshell, our outlooks for growth on the electric vehicle market is still strong for Mersen. By 2029, we should go beyond 100 million EUR in turnover versus 30 million EUR this year. That means about 25% annual growth each year, confirming the market's potential for us over the medium term.
Thank you.
I'm now at the end, I believe. And now with Luc, we'll tell us more about a buoyant market for the group, the SiC semiconductor market. Pardon.
Oui, deux mots. Thank you, so this is a term you hear quite a bit. It's really about converting power while minimizing losses. The current that comes to this room probably was converted about 10 times. Nuclear plants are alternating current. It comes in a power station where the voltage is lowered, and in a house, you have LEDs, you have your lights on. That is direct current. It's DC. If it's coming from a photovoltaic panel, it has to be transformed as well to turn it into DC. So when you transform power, you keep losing some of it. If it's just at a small scale, the losses are small, but at a very large scale, you lose the equivalent of several power plants, nuclear power plants. So we have new applications.
There's one that we've been talking about a lot, which is silicon carbide, which is used in power conversion. And again, they're not used in iPhones or computers. They're really used only for power conversion. And more recently, there has been this new application, which is electric vehicles. So the product was initially used for LEDs, blue LEDs. Wolfspeed started. We started to do business with them 25 years ago. They were using silicon carbide for substrates. Then they were used in industrial applications. And then in 2019, half is used by the industry in different applications, industrial applications, and already 50% in electric vehicles. The 50% that you're seeing here is 100% Tesla. This chart shows that this is a business which is very, very young. It's only starting. So only with Tesla. Here are the numbers. This is when Tesla made 400,000 vehicles per year.
Then in 2023, 2.7 billion. So what does that include? It includes still a lot of Tesla. They made 1.7 million vehicles per year. If you do the math, the math probably will give you 2.4 billion. There are some Chinese who also started to put SiC chips on their vehicles back in 2022 and 2023. And I'm not sure there is a vehicle. I think the first one was Porsche. But otherwise, there are none in Western Europe. So what you're seeing here is not phenomenal. It's just realistic. It's not a dream. We don't know if it will be in 2029 precisely. But the market after that, by 2035, is even more massive. What you have to keep in mind is that in 2023, there were not that many brands that were using this technology, just Tesla and a few Chinese brands. And that's it.
The number one benefit of the product is greater efficiency, fewer losses. And with the losses, you can either have a smaller converter as opposed to a silicon converter. That's what they do in the automotive industry. There are silicon converters, and then you have SiC converters. They're only one-third in terms of size. And they consume less power. You have fewer losses. And you have a vehicle that can either have more autonomy or, if you have no space, you can reduce the size of the battery. You could put a smaller battery on a smaller vehicle. So you can really use it in a very flexible way. And so that's really something that is decided by manufacturers. And the more voltage, the higher the voltage of the battery, the more attractive the technology becomes. So 800-volt batteries should all be equipped with SiC.
800-volt can also be charged very rapidly. You can also save on copper. I think if you do the math, between 400 and 800, I think you can save 15 kilograms of copper on a single vehicle. You have less voltage, less intensity if you're interested in the technical details. So all in all, it's a great product. It's a great product, but it is expensive. It takes half a wafer, which you saw outside, to equip a vehicle. The forecast is one wafer that will equip five vehicles in the future. So let me just rewind here. So the 10 billion include the fact that it's half a wafer per vehicle. But in the future, one wafer will equip five vehicles. So these guys, they did the math. The only unknown is how many vehicles will be sold then, that we don't know.
We do know that there will be many, that many will be sold. So these are the benefits of SiC. You've also heard possibly about GaN. GaN can be used on a SiC substrate. It will be a slightly smaller market. You might find some of them on charging units. So this is a different slide. What this shows you is the transistors. These are some of those you saw outside. You have the black box with the one that has all these chips. You can find those in charging units. Alpi tronic uses SiC transistors. You can use cars that have built-in chargers, which means you can charge anywhere. You have an AC-DC converter or DC-DC converter, depending on the voltage. You have SiC MOSFETs. And in the main converter, you have this technology as well, the one that turns the 800-volt DC into 600-volt alternating current.
In the car's converter, you can't remember the amperage, but you can have up to 40, so I guess in total, you have between 50 and 100 on an 800-volt platform of these SiCs. One wafer in the future will equip five, so we're reducing the costs with high-performance products. Crystals are also better, and then some companies make those modules, these black modules. There are some plants now that are fully integrated that make those black modules fully automatically. The costs came down quite substantially, so some say, looking at the cost of automotive, that the product could be used in other applications in the future as well. We haven't really looked into this. If it's the case, it would obviously be good news.
Mersen is not visible in the final product, yet it plays a crucial role in the SiC power semiconductor manufacturing process.
In today's world, and even more so in tomorrow's, SiC power semiconductors are at the heart of a number of equipment in renewable energies, electric vehicles, industrial drive, and motors. Without these hundreds of tiny components designed to switch electric currents at a very high level, nothing would work. Now, SiC semiconductor manufacturing is terribly complex, tough, and meticulous. From SiC sublimation growth to cutting of wafers into chips via epitaxy and ion implantation, each stage is subject to performance and cost challenges. Within this manufacturing process, the quality of the material is key. Leading expert in Advanced Materials, Mersen supplies fine-grade graphite and thermal insulation for furnace linings, wafer carriers in graphite for epitaxy, and electrodes in graphite for ion implantation.
Our strengths: our ultra-highly purified graphite and high-performance insulation with exceptional qualities, thermal resistance at SiC melting point, ability to prevent contamination, resistance against aggressive media, ability to machine intricate and precise design, customization at every stage. Thanks to this magical material that improves efficiency and ensures top-quality SiC substrate, Mersen is a key partner for SiC wafer manufacturers.
[Foreign Language] . Okay, this video illustrates that it is a quite unique material. The tooling stage or step is important. The purification as well. It's not the hardest one, but important. The coating of parts, hence the investments we made in two or three of our plants to do all this. Everything you saw in the video, the outside is a felt, which requires no specific interventions. But the composition and the texture is such that we have very high insulation capabilities.
Wolfspeed uses a highly technical product for its insulation. We're the only company that provides it to them. And they are really the only ones to use it. A quick word on p-SiC. The partner is Soitec. So this is the timing of the original product. There is a bit less pressure now to deliver large volumes for next year, which in a sense is good because we're still fine-tuning. Please keep in mind that SiC or p-SiC, the wafer size changes in millimeters. The original size is 150. It's the wafer which is outside on the table. And now we're going to 200 millimeters because the equipment is 200 millimeters. So it makes a lot more. It's a lot more profitable to go for 200 on existing equipment. So we're working very actively. You may think it's a detail. It's absolutely not a detail. It's actually quite complicated.
But so far, so good. We're sampling 80% of what we're doing today is 200 millimeters. If STMicroelectronics goes for it, they will go for 200 millimeters as well. You have more chips on the same component. And yeah, that's it. The beginning of 2025, we'll be getting the final equipment, industrial equipment that will be used to deliver the volumes we want. So we're going to have to use those new machines, which we will do starting in 2025. We're not talking about massive volumes for Soitec. We're going to continue qualifications and certifications with customers in 2026. We will start delivering and more so in 2027 since the plan was the SiC, the PolySiC plan was slightly postponed. So this is where we stand. We have the Gennevilliers site. The biggest challenge, I think, here is for STMicroelectronics to use it.
I think really that would be the bulk of the business. I think they have 35% or 40% market share in SiC processors, followed by Infineon, but they are really one of the biggest players. Okay, here is a quick animation to show how this works. Mersen's right in the middle, supplying the graphite equipment for wafer manufacturers. We have SICC. SICC is a great company, a Chinese company. We've been in business with them for 10 years. Some Chinese companies have been quite active. We have newcomers as well who are less advanced, but SICC is qualified with Infineon, Bosch. They are a great company generating good sales for Mersen. Their wafers are used by Infineon. I think they are shipped to Infineon Germany and possibly in Singapore or China. Then they deliver to Xiaomi. After making phones, they now make vehicles. Great vehicles, by the way.
There is Wolfspeed. Wolfspeed, they deliver a lot to STMicroelectronics and eventually Tesla. And then you have all the other SemiSiC, also a Chinese company, Coherent, onsemi, SK Siltron, SiCrystal. So they deliver to Bosch and all these others. Wolfspeed has been not still manufacturing at full speed, even if their wafers are doing really well. Hitachi as well, ROHM as well, and car manufacturers. I think this morning or yesterday, I saw that Stellantis is also working with Infineon. And I think STMicroelectronics also announced a new partnership, a new deal with Renault, Renault cars. So all these companies really want to be certified and qualified with car manufacturers. And as you can see, we are right in the middle. Somebody asked me the question whether we worked with Chinese companies and BYD.
BYD at this point, they're even trying to develop transistors. They want to be fully integrated, but BYD right now is buying transistors from Bosch, and the wafers were probably made by Wolfspeed. So that means that includes us. Okay, moving on to the next slide. Let's rewind again back to 2022. This is where we were asked to sort of look forward, starting with 2022, but also beyond 2023. We didn't quite have back then what it took to deliver. When all these people started to ask for graphite to address this, we were able to meet only 50% of their demand because they came in with massive demand. We said we can't. It would require too much investment. We should have invested double, EUR 600 million to deliver what Wolfspeed had in mind. So we did what we could. We did our very best.
The market eventually wasn't asking for that much. You can see the growth peak in 2023. That was for platforms that were going to be launched in 2024. Platforms started to be postponed. That's where we stand. In 2024, people were honoring their contracts, but the market started to shrink. There was a real peak in the industry in 2022, 2023. Now we don't know what to do with the transistors. We have this stock of SiC transistors or converters with our customers or elsewhere. We don't really know how many there are. Indirectly, in a sense, we know because when we talk to our customers, we think there's a scenario. We have a realistic scenario that shows that we have to postpone everything by three years. You have the original scenario and the updated Mersen scenario.
This is based on numbers we've collected from our customers and from the business. This is not something which is developed by consultants because there are no consultants who know the business the way we know the business. Again, putting things a bit in perspective, things are going to pick up quite significantly, we expect, and we have numbers that show that in 2030, we'll have to see exactly where we are and where we will stand in 2030, but 40 million EVs produced, including in China in 2030. The world is only making 90,000 right now, so if you need that, that means you have to produce a lot. It's not a business that stopped. It's simply a business that took a bit longer than initially planned, so in the short term, we're short EUR 100 million in 2024 because of this decline in demand.
The target is still sales above EUR 220 million by 2029. This includes some Soitec with a lot of work in the 8-inch transition. There will be a lot of technical progress as well. We're really trying to help our customers become more productive too. Thomas will come back to investments. We can postpone some of our investments, but not all of them. The available capacity will not be used at 100% next year, and our investments are not dedicated to SiC. They could be used in other applications too, not just by snapping your fingers overnight, but in the long run, definitely. I'm convinced that we will sell graphite, which was initially to be used for SiC in other areas. Olivier, I think this is where you're supposed to be back.
Right, so I'm going to tell you about semiconductors and silicon. This is not SiC.
This is a much bigger market. On this chart, as you can see, you have a forecast by Applied Materials. They're a major player in the industry. As you can see, it could reach over $1 trillion by 2030. Growth is strongly driven by two factors. First of all, connected devices. Devices we use in our everyday lives, they require storage, memory, and there are needs for growing computing power to drive the development of artificial intelligence, AI. So this tremendous growth can benefit Mersen because we are positioned at different key stages of the manufacturing process. First of all, the high-purity production for which customers use are graphite and rigid insulation systems resisting very high temperature with very high levels of purity. As you were able to see next door, if you're physically present here, we have an oven example.
We also have an active role in epitaxy, which is the deposition phase on wafers. Once again, next door, you were able to see that we make carriers that can be used to carry those wafers on which we then put the active layers to make semiconductors. The same thing for the ALD deposition phase. Our customers use those carriers to deposit one atom after the other, which is critical if you want very fine structures like one or two nanometers. They are absolutely necessary to densify the amount of memory per square centimeter on a wafer, for example. Last but not least, ion implantation. We provide graphite parts that are very, very fine with very high levels of purity. This ion implantation allows us to change the physical and electrical properties of silicon wafers.
So this market looking forward could exceed EUR 100 million for Mersen in 2029, which is double our current numbers. To do so, we're going to leverage our local footprint. As you can see, we have a true global footprint in all geographies to serve local markets, to address growth, be it in Asia, in the Americas, or in Europe. So Mersen's perspectives are good in the semiconductors market. Back to Thomas for the financial part. Thomas.
Thank you, Olivier. Before we talk about finance, what's true about renewable energies and also true about silicon semiconductors is that we could actually be well beyond what we just did. So EUR 100 million is really the bare minimum. We're being deliberately cautious. Our commitment is EUR 100 million. So let's move on to the financial part of the presentation.
Once again, I would like to show you very quickly the company's historical financial performance. This shows where we have a robust profile. We will tell you again about the postponing of the plan, and we'll again insist on the fact that we have a robust financial profile. I'm not going to spend too much time on this slide. Luc showed that our sales increased quite substantially compared to the past year's EBITDA. Same thing as Luc said. We've improved our EBITDA margin, 60 basis points if you compare 2021-2023 to 2016-2019. And if you apply it to sales, increasing sales, where you can see that you really have the double effect in terms of EBITDA value. And the volume effect has been essential. This is really what is driving us.
If you believe in the growth of the market, well, then you can believe in us and in our EBITDA. The prices, inflation are such that we can cover our cost increases and inflation increases by better productivity and pricing, as we have demonstrated over the past years. The good reason, once again, behind this is that since we have products that are tailor-made and that are co-developed with customers for years, every year, they need to be upgraded. So customers just don't decide to switch vendors overnight. They don't do this. Even if there is a price increase, if there is a 5% increase, it would be too complicated. I wanted to show you the profitability of our two divisions. This is the recurring operating income. You have Advanced Materials in light pink and Electrical Power in orange.
You can see that those histograms show progress between 2021, 2022, and 2023. Again, this is the operating income. Looking at the margins, comparing 2023 to 2021, we have a 1.3% increase for Advanced Materials, 1% for Electrical Power. The trends are not necessarily the same year after year. We had a real increase, a real peak with Advanced Materials between 2021 and 2022. Then things became a bit more stable in terms of margin because we had more amortization. The Electrical Power division made more progress in 2023. You can see that there is a profitability gap, which is quite significant between those two divisions, 5-6 points. An important gap. In terms of ROCE, the gap is actually shrinking. You can see that there is a two-digit point gap between Advanced Materials and Electrical Power.
This gap was actually even smaller in 2022 because there are two different financial models. Advanced Materials has products that deliver better margins, but it is a division that is more capital-intensive than Electrical Power. So in terms of return on capital employed, this is fairly consistent. The different models, but they also have a lot in common. First of all, they address the same customers. There are synergies, marketing synergies, commercial synergies between the two. When you develop a new product, for example, when you're entering a new business segment, I can give you one example with solar. Both divisions work in solar, but we started with Advanced Materials. They developed a true customer and market expertise. And thanks to this expertise, Electrical Power was able to position itself with a new product on the downstream part of this industry.
They both share industrial sites, not in all countries.
In the U.S., it's quite mature with dedicated sites. But in some countries at the outset, we always start with what we call multi-activity sites, accounting for about 20% of sales, 20% of sales in sites covering both, enabling us to develop faster. Let me take an example, India, a recent example. It has been there in India for some years now, Electrical Power, part of its clients in India. But it was far simpler to set up operations and install equipment in India because we had a base for Advanced Materials. So it went very quickly. It was amazing for us to be able to produce in India entirely. In India, we are probably at 80 or 90% of our products manufactured locally. Five years ago, for Electrical Power, it was very little. And of course, at group level, we have seen that certain functions have been pooled.
R&D, as Luc said, and innovation. But you can imagine many other functions we like to pool. Overall, a cost base of around EUR 75 million-EUR 90 million that is shared between both divisions, and that's something. Moving on to cash. Luc also said this. It was great for cash between 2017, 2019, and 2023. We've always been cash flow positive. So we said, fine, you have negative free cash flow. Well, for two or three years, when you invest what we invested, of course, you have a negative free cash flow, but you'll see. That was to create more value, ultimately. So in particular, I wanted to show you this free cash flow, EBITDA minus CapEx, giving you an idea about our net free cash flow. And this is for traditional markets that account for 70% of our revenue.
All industries, traditional. Well, may sound negative, but all the industries that are neither semiconductors or EV or renewable energies, it's all the rest. All the rest, that means constant large cash flow that even increased significantly in recent years. Why? For several reasons. You don't need to invest heavily. We have a fabulous customer base, highly diversified. Luc spoke about businesses that couldn't be recurring, but there are others. Great resilience and ability to increase prices because we're talking about tailor-made products, small amounts, so you can increase prices easily. And we have a rigorous management of working capital requirements and cash. I spoke about that to come back to 2022. At the time, we faced an unprecedented situation using 100% of our production capacity, so we couldn't produce more. At the same time, our customers said, well, we have a huge demand. So why did we invest?
Because of that demand. That was significant because we had the capacities and we had to invest anyway because we were at full capacity. Plus, you saw that we can afford to invest because we have a significant cash flow base in our traditional markets, and we could secure a part of volumes through long-term contracts with customer commitments, so we invested, and today, we're here with a situation that is not quite the same. As we said before, with market outlooks, we spoke about the SiC semiconductor market, which is lagged, and excess stocks for semiconductors on solar as well because there's a lot of production. We spoke about that at great length in October when we released our revenue. We launched an investment plan. As Luc said, it takes a long time to produce this equipment for infrastructure. It can take a year or more.
So we spent today what we ordered a year ago or even more. This is why we can't stop investments suddenly. Plus, there are geopolitical tensions that have been heightened, as you have seen in the media, and we know that. So what are we doing in light of this situation? First of all, and here I'm speaking about investments in our growth plan, we reduced what we could, reduced by 30-40 million EUR versus the initial plan. Since there was a bit of inflation, all in all, that's 280-290 million EUR between 2023 and 2025, as compared to 300 million EUR that we estimated at the outset. That's always, again, it's not a lot, but as I said before, we were strongly committed.
Looking at investments, but overall this time, I'm talking about investments, growth, plus investments in the growth plan, plus our investments for all the rest. You also have growth, but we can call it our normative investments. You can see where the peak in 2024. 2024 was very high. Good news, it will slow down in 2025 a bit, 2026 even more, and we will reach a small amount of revenue. I can see analysts in the room. Let me just say that don't take the figures down to the million. It's just a range, just to show you the trends. So we reduced investments. The other thing we have been doing is adapting. As Luc said, we had several plans each year. We'd say that this plant isn't great. What can we do? We tried to help it to recover, and here, we accelerated.
We shut down an unprofitable plant instead of trying to increase its profitability. So we announced a certain number of measures, as you can see on the slide on the left-hand side. We announced this in October. What we added since then were two things.
The shutdown of a plant in Mexico, bringing back businesses to the US, and a transfer of EV fuse production. We're looking, keeping some in Mexico, but most of it will be in China, to have a critical mass that will be more significant. And so that will allow us to gain a bit more. We announced EUR 15 million in EBITDA, I believed. We're now at EUR 17 million . It will cost a bit more. We announced 12 million EUR in the cost of cash, EUR 20 million overall. It will be EUR 14 million in cash and EUR 23 million overall, EUR 23 million in this plan.
That will be in P&L largely, but most of it in 2024, whereas the cash costs will be more in 2025. So you can see, I don't like to talk about a payback, but it will be fast when you reason in terms of cash. It's EUR 14 million in cash cost for EUR 17 million in EBITDA. I speak a lot about EBITDA, why? That's because, as Luke said earlier, in 2025, it will be a year of transition, and we have amortization that will be coming in our P&L because of our investments, about EUR 10 million to give you a rough idea. That will come, whatever happens. What we are focusing on is on EBITDA. That is what creates value and generates cash, so that is why you will hear us speak about EBITDA more and more.
In addition, because we are limited when it comes to the ability to reduce our investments, we decided to launch a plan on inventories management. We took our 10 biggest plants. We looked at what we can do in detail with new ideas to enhance things, and we'll gain EUR 30 million in inventories on a like-for-like basis. Maybe, well, one year may be a bit fast. I don't know. Maybe between one and two years. Two years could be long, and one year could be slow. Setting aside the plan, I spoke about geopolitical risks. Luc spoke about our industrial footprint, which is amazingly strong for us and being close to our clients, developing new markets, being close to our clients above all. Because for these products, you must be very close to Samsung to offer them innovative products, to take that example.
At the same time, it's also a major asset to reduce risks, risks that are geopolitical primarily. We've spoken about customs barriers. I'm not saying we won't be impacted by customs barriers, but not very much because we produce where we sell, and we source locally to, let's say, 80%. So we're immune to customs barriers. So if we have customers, of course, our suppliers, who they themselves are impacted, it could impact us indirectly. But as you know, our biggest client is 5% in revenue. Our biggest supplier is 1% of our procurement, meaning that we are very diversified. Secondly, in certain times, geopolitically, exchange rates shift a lot, and we have no foreign exchange effect on transactions, on competitiveness. So we produce where we sell.
Of course, a foreign exchange impact, but in terms of conversions, typically this year, we're losing about EUR 10 million in revenue because when you convert everything into euros, there is a negative impact of foreign exchange, and that has leveled out. Finally, we have debt in local currencies and cash too in local currencies. So our balance sheet is also protected. It's not totally insensitive, but largely insensitive to foreign exchange fluctuations, especially when it comes to the financial structure. Coming back to the plan and the original idea, it hasn't changed. Our idea was to almost double EBITDA between 2022 and initially 2027, now it's 2029. Still the same figures, but doubling EBITDA. I took the guidance for 2029 to set 19% EBITDA for EUR 1.7 million. Looking at the free cash, we plan to double it by 2.5 times between 2022 and 2029.
It hasn't changed, but it is true that there'll be a two-year lag. The debt. As you can see, the debt increased significantly between 2023 and you will see at the end of 2024 why. We expected that because we have EUR 220 million in investments, the peak, and we add to that EUR 65 million in acquisitions in euros, plus earnouts in the acquisitions. The debt will be roughly a bit higher than EUR 400 million euros. Of course, that's a significant change as compared to the previous period. This is why we intentionally gave you a long bar chart. When you look in terms of financial structure, our financial policy that you have been hearing about for years and years is between 1.5 and 2 times leverage. It's been 13 years to have been below 1.5, and you can see we're just borderline before.
So we will come back in our financial policy. We announced between 2 and 2.3 in leverage at the end of the year. So we're very far from our financial covenants at 3.5 times, our bank covenants. Luc said I'll talk about acquisition, so I will. Okay. Next slide. We announced about 100 million EUR in acquisitions in our plan, as you may recall. Our plan for 1.7 billion EUR. From 2027 to 2027 to 2027, we looked at 200 cases. Thomas Farkas is in charge of acquisitions and strategy. He worked a lot on it, so he acknowledged that. 200 cases. If you look per year and taken away 2020 and 2021, which were special years, that means about 30 cases per year, more than 30. Rigorous management because at the end, we had three, three this year. What did we gain from that?
Family-run businesses, companies offering us new expertise, adding to our geographical coverage or expertise. So mainly expertise, a bit of consolidation, GMI and Barlow, a lot of consolidation, bringing in synergies. The C&K acquisition. The three acquisitions were in the U.S. Two important things to note. We didn't pay a lot for them. Five times EBITDA at a reasonable price. Once again, a very rigorous management of our acquisitions. The second important takeaway is that for the time being, we have no other short-term projects. No short-term projects because we know that the valuation is low. When you make an acquisition, you also look at multiples. Today, we won't look for multiples of eight times EBITDA with our current valuation. Normally, no acquisitions to be expected in 2025. To be comprehensive about the financial structure, we need to speak about liquidity.
This slide's at the 30th of June, EUR 300,000 in availables and EUR 100 million in cash. That dropped, of course, at the end of the year, but it will be very abundant. We still have a policy of excess liquidity, and we'll have a lot of liquidity to manage our investments, even to refinance the Schuldschein in 2026 for sure. So we're already working on refinancing because we still have a policy of excess liquidity. That has always been the group's policy. It costs what it may cost, but in a group like ours, it's always good to have more liquidity rather than not enough. So to conclude, I can say we have a value-creating value model. I hope that you're convinced of that. We have a long-term plan that is intact, hasn't changed much, but with a lag.
So there are one or two years that will be years of transition, well, anyway, 2025. We'll take steps to adapt and to accelerate on EBITDA and cash to make a bridge over this year of transition. I think, Luc, the third time for me.
This is the third and last time, I believe. Okay, coming back to the group's potential, I think we've said a lot about it. We've covered a lot of outlooks above and beyond 2027 and 2029 with markets that will continue to deliver. To repeat roughly what we said at the beginning, the figures shown by Thomas show a gap, a lag. There's no mixed effect. The price is no surprise. We just shifted the figures by a year or two versus our plan. Let me add one. We should also look towards 2029, roughly.
There are sectors where it's a bit complicated to identify revenue today, but we can feel that the group will be able to grow. We're talking about for nuclear energy beyond 2030, even if we're already working on it today. As we said before, we have prototypes. It's now that we must position ourselves. If you don't have projects and plans, you won't make it. We're looking at many projects. We know that many won't go very far. We know that. In France, there's a pre-selection of projects in many different areas that started a year or two. So we already have projects that will not receive support from government or technical support. But we are not sorting, are not selecting at this stage, not screening. We answer calls when required. Smart grids, coming back to nuclear. We'll see.
It's already a bit complicated to transform that into figures, but it gives you revenue, and this will grow. Coming back to nuclear projects, we spoke about that in the introduction in SMRs. These are small modular reactors. It may make you laugh, but the goal is to put them where needs are in a plant or in a village. We spoke a lot with people who are pushing for them. We need to work on acceptability by the population, even to put it in a plant. I'm not sure that you'll sign to have that in a plant in a remote area to generate energy. But you must know that these are technologies. To move away from Hiroshima is a bit far, but it's very secure. People work on systems. I'm not saying it's risk-free. You always have ways to evacuate and fuel to put in the machine.
But in principle, they can't create incidents such as what we have seen with Chernobyl, for example, or in Japan. But nevertheless, it's nuclear. I think people working on it have a lot of work on their hands to explain it to the population for it to be considered as normal. Technically, it's very interesting. There are concepts that work with waste, mountains of radioactive waste, and we don't know what to do with them. We store them, and they're consumed again. SMRs can operate for a thousand years, and there are other principles using fuel that you need to produce, but they're more sophisticated than with EPRs, or they consume less at any rate. Just take away two acronyms. If you read any papers on this, we can have things to do in HTR or HTGR. They can be bigger.
We're talking about SMRs, but there's an HTGR in China for 1,000 megawatts, but in a small one with 20 or 40 megawatts, you can have 100 tons of graphite. Sophisticated internally, less externally. If you talk with people who are very optimistic and very motivated, some believe they can make 100 per month in 2025. That's 100 times 60 tons, 14,000. Just to show you what this entails. We feel very secure about this, but we produce prototypes following this closely. For SMRs, there's a lot to be done. In the US, well, lots of things in France. Jimmy may not be familiar with it. On HTGR technology, we also have silicon carbide with prototypes made. The project, that's how it works anyway. Some companies will develop and pass away, and others will make it in 2035. Some will be selling SMRs.
So in Europe, in France, and a lot in the U.S., four or five projects in the U.S. between military and civil operations, so that's it for what we call smart grids or microgrids. Not quite the same thing. Smart grids are smart networks where you put in a lot of smart systems measuring consumption, forecasts on consumption, adjusting how power plants adapt to demand. That means for us making more and more products that communicate. We already have offers for Germany for smart grids, sending information to operators to see if something has broken down or needs to be changed or to measure current or power. I'm looking more at microgrids because microgrids are pretty nice, in fact. Small local networks that can supply small localities or a hospital or a village in the middle of nowhere.
It's a good concept because there is production using solar panels, often storage for nighttime, and supplying houses with DC current, lots of conversion. We see lots of examples with brand new players proposing after investing turnkey for people living there. They invest in green energy. People love green energy. So it's not bad. We see an interesting future there with going faster than nuclear. And we have a wide range of products that could use these applications. So to be monitored as a growth driver outside of the scope we presented this morning for Mersen in, let's say, around 2030. To conclude, let me repeat what Olivier and Salvador and Thomas said about the group's robustness. It gives us the wherewithal. We have no issues in terms of finance or in terms of layout.
We're not going anywhere to ask people how to set up an industrial operation in India or Brazil or in Pennsylvania, well, in Pennsylvania, we have three anyway, so we're ready for any opportunities. As some, we're following them in India. We did that in one year. People are still wondering in France about setting up operations in India. Apparently, it's complicated, not for us, and we produce, and there are interesting outlooks, so we're ready, more than ready for a rebound in the business cycle. All of this about ups and downs for electric vehicles. I would imagine that it would be a bit more structured when it started. It all has to be well organized because of what's at stake, but there are ups and downs. We will absorb that. 2025 will be less good than in 2024, but we will make it.
As you have seen, later on, it picks up rather quickly. So let me just repeat that these effects will just postpone our plans by two years. There are still groups that are 100% in the automotive market that are suffering. That's not our case, but we will bridge this difficult period. Just to say this is a fine company full of a bright future. Véronique is leaving. No, that means I need to stop talking, I guess. So thank you.
Of course, we'll now move on to a Q&A session, starting with questions in the auditorium. For those of you on the webcast, you may ask your questions, and we will take them at the end of the questions in the auditorium. No questions. I can't imagine that. It was crystal clear, then. Thank you for the presentations. I have a quick technical question.
I'm not an expert in semiconductors, but you spoke about the SmartSiC technology. Soitec , I believe it's called. From what I understand, in the SiC, the price of the substrate counted a lot in the product's profitability. I believe that this technology allows you to divide the cost of the substrate by six. I don't know if that's the case. It's been a long time that we have been waiting for a contract with ST several quarters now. What are the hindrances for obtaining that contract? Thank you so much.
Well, in fact, I don't have any direct information from ST. I think it took them a good year putting SmartSiC on traditional lines for their transistors. They did a lot of fine-tuning as well. The product is not developed quite the same way as traditional products. There are some differences.
Once you've made 1,000 or 2,000 transistors and you've double-checked everything and you're sure it works, you need to put them in a converter and then place them onto a vehicle. It's not something that happens overnight. I said 1,000 or 2,000, maybe it's 5,000 for the qualification batch. Each manufacturer, you need to do this for each manufacturer. So it doesn't happen in two or three years. And then with STMicroelectronics, as you know, they buy wafers from Wolfspeed. They are also trying to develop their own. They have their own ovens in Catania, 200 to meet their own needs. And I think that eventually what they would like to do is to have their own SmartSiC. And I think they will know more once they've reached some volume, the technical strengths. I don't know exactly what they will be.
I don't know when the signature between Soitec and ST will happen. It's actually none of our business, but I don't see anything picking up before 2026. We're still very much in the developmental stages, and it's not that obvious. And in theory, we should be the ones delivering the p-SiC, which plays a key role, obviously. If you take a mono wafer, if you split it 10 times, so you have 100. If you split it 30 times, it's 30 times less in terms of cost. So we'll have to see what Soitec can do. We have already worked on our pricing, our costing. It's obviously not something that comes free of charge. So we will see. We will see with them and with transistor manufacturers how we position the product.
I think we're probably going to go for 200. 200 wafers are so expensive that I think there will be more opportunities for 200.
Hello, Jean-François Granjon. I have a couple of questions. I understand why there is this lag. Obviously, you're not expecting a major contribution in 2025 and 2026. You said it's going to take time. The contributions are going to come gradually. Why two years when we don't really see the strong contribution of these new markets in 2024 and 2025? Do you think the other businesses will also be more limited? I'm really trying to understand why there is this postponing of two years. That's my first question. The second question is about the contracts. Two years ago, you told us about the contracts. You said 2027 is important with Wolfspeed. And ACC, you revised the sales of the ACC contract.
You adjusted it since. What's the status of those contracts? How about their size, their duration? Is any of this questioned? Any postponing there as well? And question three is on CapEx. There's been a peak in 2024. There will be less CapEx in 2025 and 2026. You mentioned the negative free cash flow in 2024. Will there still be a negative free cash flow in 2025 and 2026? Thanks.
Contracts are postponed. The global volumes are not questions. The forecasts for 2025-2026 are lower. SiC deliveries are going to increase in 2025 and 2026. 2025 was not going to be a bad year. We put a lot of money, and we invested a lot of money to deliver growth in 2025. And obviously, it's not going to happen in 2025. It's not going to happen in 2026.
You saw this on one of the slides.
We don't know what will be the rebound. All we said is that we are somewhat postponing. We're postponing the forecast by three years. We don't know what the stocks are. We don't know when stocks will be required and when supplies of SiC will resume. So we've taken a normal scenario. If all countries decide to give subsidies and make the products free of charge, then there would be a peak, but I don't think this will happen. So I expect things to resume in a quite standard way, and it could happen maybe a little earlier, but I don't think so.
Contracts?
I think we answered, didn't we, on contracts? Well, actually, right. There is ACC. At one point, we were told that the platform we were on would be bigger than the other one. Now we're back to the initial contract with the initial volumes.
The vehicles are the 5008I or E, sorry, which is going to be produced at the end of Q1. 3008 as well. There will be additional vehicles on which we were not planned, which Stellantis is also developing with Chinese packs that are available.
We're simply waiting for their manufacturing capabilities. They seem to be having some yield problems, but this is not a problem. We've seen the same thing with Soitec, so we know what it's like. The program for next year is to really accelerate more in H2 than in H1. I don't think we will have the numbers we had two years ago. As far as free cash flow is concerned, I think you include everything, taxes and so on. Being positive in positive territory in 2025 is going to be a challenge. The main negative free cash flow is 2024.
We will talk about this again in a more detailed way in March. I'm being deliberately unclear because I don't know what else to say at this point.
Hello, Jérémy Salé. I'm with BNP Paribas. I have two questions. I understand that on two of your big markets, 2025 and 2026, will be years of transition. How about the 40% remaining on other markets, more buoyant markets? Are they also going to be going through a transition, or do you think that they will continue to grow at a pace of 2% to 3%, which is what you said in the past? And as far as M&A, you said 100 million EUR until 2027. How about 2028 and 2029? Will there be any acquisitions between 2027 and 2029? Well, we put mergers and acquisitions on hold, as we said before, for an undetermined amount of time.
If the stock price picks up again, if SiC all of a sudden starts doing better, then we might reconsider this. So I can't really answer the question. We're not looking at any acquisitions for the time being in 2025. Right. He said no, that's no.
Your first question was on the other markets. You mentioned 40%. We have a little over 25% on those initial markets. We have no doubts about those markets. I'm not going to give you any guidance at this point. We will come back to this. It very much depends on the overall economic context. It could be a bit more, it could be a bit less. Honestly, we don't really know at this point. And I think it wouldn't be the right time today to discuss this.
Hello, I'm Steve Vaughan. I have several questions.
I'd like to come back to ACC and the outlook between now and 2029. In the initial plan, we had ACC, and you were hoping for potentially two other contracts to meet your ultimate goals. ACC, they canceled. Now they're back to one. So things are quite clear for 2027. But looking beyond 2027, I have a fundamental question about the development of batteries in Europe. What we have is nickel batteries. We know their performance, but these are old technologies compared to what the Chinese do. The Chinese are already developing nickel-free batteries with pure lithium batteries and dry batteries. And Europeans are really lagging behind. So the question I have is, will there be any other projects besides ACC? Or has Europe simply lost the race? So hence my question, 2029. You're not active in China on the battery segment. You're only a supplier.
How confident do you feel about this? So this is my first question. Maybe you want to answer that one first, and then I can ask the second one.
It's a difficult question. I think Northvolt, they were going to make packs for Volkswagen. Volkswagen took a stake in the company. The other gigafactories outside of China, there is Envision, CATL too. There is a Korean company that has one site. But a year ago, we had too much in the pipeline, too much food. And plus, we had this postponing phenomenon. I don't know how far the Verkor project will go, the Envision France project. We have the semi-solid Taiwanese plant, which is not finished yet. So everything is taking more time. If European manufacturers bought Chinese products and would then assemble them, they would then be able to use our bus bars.
The problem is there are no European gigafabs. The Europeans buy packs developed by the Chinese. They're sealed. They're ready with all components. In that case, we wouldn't be able to add anything. We've had very few discussions with companies like CATL and the Koreans. And these are really the products that are on electric vehicles that you see on the road. So ACC, they were asked to switch to another chemistry, chemical for us, whether it's nickel or something else. It doesn't really matter. It requires bus bars. Again, I think the Chinese are really ahead. I'm not super optimistic about European gigafactories. We don't really expect to have bus bars everywhere. Even at the best, the contract, the ACC contract is EUR 40 million. It's not nothing, but it's not a huge contract. It's still very low.
We think that by then, we can pick up other things to compensate. We have one or two bus bar contracts for converters that came quite unexpectedly. They're relatively small, five or six million EUR contracts. Anyway, ACC is important for us. Even if the contract were downsized, we'll have to see what we can do on battery bus bars. If the Chinese have everything, well, then they have everything. But who knows? If we have a new technology that uses our technology in Europe, it would be interesting. In Europe or even in the U.S.
Technologies keep changing. The bus bars that are outside here and the ones we're currently developing are quite different. We're keeping our efforts. We're continuing our efforts. Are there any American projects? You talk a lot about ACC with Trump. In theory, there could be American projects.
Absolutely. In Europe, we know there are some constraints, as Salvador said. I think American manufacturers are probably happy to continue and sell internal combustion engine vehicles, except for California. So everything's taking more time. There are a lot of gigafactories in the U.S. They might even have to be closed. If they're closed, it's not good. I think there are six or seven in total. But they don't use any American technology. There's two Chinese ones. Don't know about those. But then there are a lot of Korean gigafactories. And their packs are very good and more expensive.
Okay, I have another question on silicon carbide. You're selling to the Chinese to one Chinese company. The best one. Okay. So my question is the following. Do you think you will benefit from the growth of the Chinese electric vehicle market? Or do you think that eventually you could become 100% Chinese? Do you think you're in a position to capture some of the Chinese growth? The Chinese industry is continuing to grow to meet its domestic demand. How exposed are you really to that Chinese growth?
Well, once again, the equipment we developed for SiC, remember on the chart, I think there are some SiCrystal, STMicroelectronics. They make also transistors. But with products that are made by us, we don't really know on what vehicles they're used. Some Tesla. Wafers that are made by SICC are going to Bosch and Japanese vehicles. In 2023, when we had this peak, our products were used on Chinese vehicles. BYD, they don't really know. Right. They still have to buy. The day they can do it, well, then who knows? Maybe it will never happen. But otherwise, they will supply themselves.
Right. We don't do any business with them, so that would be okay for us. There is another company called TanKeBlue. We don't work with them. And earlier on, I did some math with Tesla. It went from EUR 400,000 to EUR 1.7 million. And we talked about the role of China.
Okay. I have two more questions. One question on solar. In 2013, you had EUR 120 million in 2012. Right. EUR 120 million in sales. It was 2011. So before, we had this very sharp decline. So last year was EUR 100 million. It could go down to EUR 80 million given the destocking. In between the two, the market and market volume exploded. We have a very high number of solar systems that have been installed around the globe. So if you take a look at the growth during 2011 and today, the amount of silicon has been divided by 10, sorry. Does that explain why you did not have that 10% growth? Or is it because you lost a lot of market share and today you are only on niche markets, on global markets?
Roughly, the market share is the same today as before with a market which itself has grown a lot. And then the question is, looking forward, what are your expectations? I'm trying to understand your economic rationale.
Well, in 2011, we did have a lot of market share. There was very little in China. A lot were made in the U.S., China. A lot of the silicon was made in Europe. And I think we had easily 50% market share. And pretty much it stopped overnight. And we had to switch to silicon pulling. They bought ovens. They started them. They invested heavily in ovens. We have grams of silicon per watt and the type of graphite they use for a watt. That, again, was divided by 10. During that period, starting in 2011, we went down to 40%. People were using graphite and ordering it all over the place, but they had no idea how to use it.
Our market share was quite good in China. Then some local players emerged. We did not want to have to develop additional Chongqing type sites. We gained market share, profitable market share. Then we decided to put a cap on solar, 100 for the entire group. We said, "Let's stick to that. Stay in high tech." The thing is, high tech's been declining. They have 1,000 gigawatts of cell productions. They're all producing full capacity. Now they sort of have to slow down.
At the end of last year, beginning of this year, and we're still in the middle of it. I think the Chinese government is sort of looking who's going to survive. There is Longi, Jinko, and all the others.
This year, we didn't want to deliver at very low cost. Some people don't want to deliver at low cost. They don't want to deliver and lose money. It's a cycle we know well. Market share is very, very low. Very difficult to calculate. But we have more sophisticated ovens, parts that are more sophisticated. On the cell surface processing systems, we also have quite sophisticated technology. If they're asking for more than 2,000 or 2,500 tons, fine, but we will not go above that. Olivier talked about growth. There will be growth outside of China where we can deliver 100%, no limits, no cap.
There is a lot of capability we can use elsewhere in India, for example, and we get a lot of orders from companies that make converters like TMEIC, Converteam, GE, Eks Energy, Power Electronics, Spain. So on electric, on those applications, we have the sky's the limit, but the cap has been placed at 100.
Okay. I have a last question on Mexico. We saw two closures in Mexico to anticipate the new Trump taxes. It has nothing to do with that. They were losing money. We had enough with it, and we said, "There's nothing we can do," so we decided to move to the U.S. Okay.
How about your competitors in electronic components? What are their industrial networks?
Mexico, a lot of companies are in Mexico. If there are new taxes or customs taxes, everybody will have to pay them.
Okay. But you do have an industrial presence in the U.S.? .
Right, It fuses, but again, we're all on equal footing. We have the same strengths and weaknesses. And I think in the U.S., traditionally, it's easier to increase prices compared to Europe.
Okay. Do you think it would make sense to transfer other products from Mexico to the U.S. or not?
Right now, we're in a wait-and-see position. We've made some changes. We've made some moves before. It really depends on how many times Mr. President Trump is going to get elected again. We have competitors in Mexico, and we have large subcontractors working in the automotive industry. They're obviously going to look at this very carefully.
Okay. Thomas, Kepler Cheuvreux.
Le point du free cash flow. On free cash flow, it's a bit challenging for it to be positive next year. What's behind that?
Because we have your CapEx, about EUR 150 million. We know your steps you'll take to cut down on savings. The tax rate won't change, but what are your assumptions? Luc said that 2024 will be harder than 2020. 2025 will be tougher than 2024. There'll be degrowth next year, but what if there is a working capital requirements increase? You will gain in inventories, but you may lose in terms of advance payments that gave you a good working capital requirements level. We'll give you that answer in March. One answer is that we're less cyclical than in the past, if that can reassure you in the room. Don't forget, we'll give you the answer to that in March. How about working capital requirements? Nothing? We'll answer you in March.
Let me first take some questions online.
So I was saying, the first question on the shareholding structure is on the Caisse des Dépôts. He's leaving now. It just so happens that the Caisse des Dépôts informed us on the 22nd of November last that it went beyond a 15% threshold decline. That the Caisse des Dépôts and BPI reported that. Because one holds shares in the other, so they have to make common reportings. The question is, were you informed of the Caisse des Dépôts' intention, and do you think it will continue to ease it up? Well, BPI are a reference shareholder. They haven't reduced it. The Caisse des Dépôts will reduce in a certain number of companies. They're not on the board, right? So you'll have to ask the Caisse des Dépôts. I don't have any more explanations than that. A second question raised by several persons.
Suspending the SM&A program, does that open up a possibility for share buyback? And what about the dividend payout policy? Would that be changed? No. About share buybacks, we never had a significant share buyback program apart from covering stock options. We need to make investments, and it's not our priority to carry out share buybacks. No, normally, no. For a dividend payout policy, traditionally, we have a payout ratio that depends on our results, 30%-40%, traditionally, I said, because that is decided by the board each time. It's a good question, though. But honestly, frankly, will we change it or not? I don't know. I would say let's wait and see till it'll be decided in March. Today, I can't commit myself on behalf of the board to say that that will be the case. Because there were some changes, some lags.
That's all I can say. It's neither yes nor no. I just don't know.
A question before we move back to the auditorium. You said that pure fine graphite wasn't the most critical phase. What's the degree of purification reached by Mersen? Is it greater than its competitors? And does that give you a competitive edge? There's spies in the room. ppm. We all do ppm in terms of purification degrees. Once you have the equipment, it's not very complicated to do the more complicated things behind that. But you need to have the right equipment to reach that level of purity. A few PPMs and no edges because the edge that's not good. One question about your six contracts. When you made your CapEx investment two years ago, you spoke about take or pay.
Could you tell us more about the orders to be placed or if you still have the same obligations and if they will be renegotiated?
We're discussing with our client. It's for sure that we won't have the maximum contracts. You have understood that. We're discussing with each client. There's a weak demand, but we're discussing.
And the discussion that also includes pricing, I imagine?
No. No. It's in terms of volume, pure volume.
And another question, maybe. In your presentation, you spoke about the fact that for using capacity, waiting for the cycle to turn around, could you use it to make other things? What are your ideas on that?
It's in the presentation. 60% of our CapEx could be used for other markets, potentially. We have capacity, though. So before we fill it up, we need to have fast-growing markets and other markets. We spoke about that earlier.
Traditional markets, that is. Our assumption was 3% over the long term. I'm not saying next year. I don't know what it will be. But it won't be 50% all of a sudden to take the capacity that can be used for SiC. Any other questions? Oh, yes. Julia.
I have two quick additional questions, a bit more technical. You reduced CapEx a bit. Given the business environment, that's understandable. What are the things that you plan not to do straight away? You spoke about the 14,000 tons that was reduced. What are the plans that you'll not carry out? That's my first question. Second question, just a clarification, because I wasn't sure about your explanations about your restructuring costs in cash. EUR 14 million cash, EUR 9 million non-cash. The EUR 14 million cash, would that be funded for in your provisions for 2024?
To be quite clear, the impact will be in 2024. Just a question. If it's cash and the expenses may need to be done, you could fund the provision?
We can't fund everything in our provisions. There are things that you're not entitled to. For example, moving costs for a factory that has to be cash. You can't fund the provision until you have moved. To answer your question, all cash costs will not be funded for in provisions, but a lot of them. After the EUR 14 million, you estimated how much that will be funded for in the provisions for 2024? Wait a second. All in all, 23. I said most of it in 2024. No, it's just because I thought I understood that it will be all in 2025. No, it'll be paid for in 2025.
I understand that. Okay.
Okay. Fine.
To answer your question about CapEx, it's a bit complex. Altogether, even us, sometimes we find it hard to figure it out. But roughly speaking, we made buildings in the city for Soitec as well. Buildings you don't stop at the top of the roof. You finish your utilities. And all of that has been engaged with utilities for power supplies, for infrastructures, for furnaces. All of that was done from the very beginning. The only thing that we can stop is with the latest equipment with capacity that was planned to be received at the end of the end. But that was quite early. So it's very difficult. For high-tech, we postponed our furnaces just in time. Instead of buying six in one go and putting all six in January 2025. Two have been installed. The other two are well advanced.
There's a time when we need to pay the supplier. The two others at the end, we can postpone that. That's the story, really. 16,000 tons. To get 16,000 tons and to really finish, we need to spend a little bit more. But later on, we managed to block that because it was modular. Yeah, for electricity, it's simpler because it's modular. Limited capacities. So it depends on how you increase in work that you can increase capacity or slow it down. That's 700,000 bus bars per year. It takes about 10 years. We have heavy equipment at a plant like St. Marys. You need to place an order two years ahead of time. That's a part of the challenge.
A question that's linked to that about costs. You will be building a factory. That entails investment and recruitment, theoretically. I guess you will revise your hiring plans as well accordingly.
Yes, of course. We adjust. We froze it, actually. We even reduced the volume, the numbers.
Well, that was implied, but I just wanted to know. That is the case. What we can't stop are with current amortizations, understandably.
Well, listen, if there are no further questions in the auditorium, I think on the webcast, we have answered all the questions directly or indirectly. Oh, there's a last question on the webcast. We were told that we were right to play the American card by multiplying acquisitions. But normally, I understand, according to the question in share price, doesn't reflect the relevance of our acquisitions in the U.S. Now, it's easy to answer that question. Anyway, well, after this afternoon's slide, it will be crystal clear. Well, then thanks to all of you.
We hope you won't wait six years for the next Capital Markets Day. So see you in two years' time if we revert to a normal pace. Thank you.