Hello, and welcome to the RHI Magnesita 2026 half-year results. My name is Carla, and I will be coordinating your call today. During the presentation, you can raise to ask questions by pressing star followed by one on your telephone keypad. If you change your mind, please press star followed by two. You can also submit a text question via the webcast page. I will now hand over to your host, the CEO, Stefan Borgas, to begin. Please go ahead when you're ready.
Thank you, Carla. Good morning. Good afternoon, everybody from Vienna. Thank you for joining us for the presentation of our 2026 half-year results. I'm joined today by our CFO, Ian Botha, and our Head of Investor Relations, Alexander Ordosch, and as a special guest, Gustavo Franco, our Chief Customer Officer. Before we move into the main presentation, let me highlight the three key takeaways from our results, from our perspective. First takeaway, our self-help measures continue to deliver on what we have set them up to deliver, and they are the main driver of the business improvement that we see in the number. Despite a soft and volatile market, these self-help measures were especially visible in our steel business.
To sustain this momentum into 2027 and even beyond, we are advancing new self-help initiatives across our raw materials plants, our refractory plant network, and driven by our digitization investments that are coming to a level where we can start to take advantage of them. Together, these measures enhance our operating leverage and will have a very significant improvement potential for the business once demand will recover. Second takeaway, our steel business, our cement business, and our non-ferrous businesses all performed well. While the global industrial projects business, the investment of our industrial customers into their plants remains very challenging from a demand perspective. We see our steel business on a positive trajectory as trade measures in some regions put a floor under soft steel demand. Support price adaptations to balance out our cost increases. The 4PRO business development activities decrease the churn of customers.
They improve the value delivery to our customers and also have a positive margin effect and some positive volume effect on RHI Magnesita. The industrial projects order book is rebounding now. The second half orders look solid, but the industrial projects order book still remains significantly below the historic norms. Third key message, we confirm our full-year guidance. While market conditions remain very volatile in all regions, actually, self-help delivers and will continue deliver well on the profit side. Our order book for steel and industrials is firm for the second half. The focus on cash flow is central to our H2 performance and to our intended deleveraging. With this, let me start to get a little bit more into the details. As usual, I would like to start with safety. This remains the core value at RHI Magnesita.
Regardless of the commercial environment or the operational challenges we face, our objective is always the same, ensuring that every employee and our contractors return home safely at the end of every working day. The first half of 2026 continued our safety culture transformation that we have started two years ago. Our digital global safety management system went live in the first quarter of 2026 and now incorporates an ever-increasing number of sites, including most of the hundreds of customer sites where our workforce is present. The adoption of this new digital system is encouraging. Employees proactively submitted over 110,000 safety reports that are quickly processed so that we can mitigate any reported risk.
While the rising number of the total recordable injury frequency, which includes even injuries with low severity, like first aid cases, reflect our proactive and transparent approach to safety, our priority is now to reduce potential serious injuries and fatalities. We call them SIFPs. Our focus remains on visible felt leadership, behavior-based safety rules, and strengthened standard operating procedures. Also many physical improvements happened in our sites and in the sites in which we take leadership and ownership at our customers' locations. Overall, we undertake practical measures designed to influence how we act and how we work on a daily basis. This improves accountability, it raises the standards across every site around the globe. We remain fully committed to our long-term objective, zero harm, no injuries. Ladies and gentlemen, let me pass to the financial highlights now of the first half.
We continue to operate against a soft and volatile market backdrop. Our revenues declined by 4.9% to about EUR 1.6 billion, they remain broadly flat on constant currency basis. It's a foreign exchange rate-induced decline. This shows these strong foreign exchange headwinds that we experience, it's not just the U.S. dollar, which could not be fully offset by management actions with respect to the top line. We delivered an adjusted EBITDA of EUR 165 million, which is a nominal increase of EUR 24 million year-on-year. If we compare this on a constant currency basis, the year-on-year improvement is above 40%, four zero. Margins improved as a result from 8.4% in the first half of last year to 10.3%, which was driven largely by the benefit of the management-induced self-help measures.
As a result of the higher earnings, our adjusted earnings per share increased significantly to EUR 1.81 per share. In line with the group's dividend policy, the board declared an interim dividend of $0.60, and we also paid the full year dividend in the first half of this year. Now, moving to our steel business. Our steel business remained resilient in a volatile market environment. Volatile it is. Revenues increased by approximately 5% on a constant currency basis, with volumes being down slightly. Self-help and our operational discipline supported these margins. Gross profit improved 17% on a constant currency basis to EUR 249 million. We successfully adjusted pricing to the ever-changing cost realities, energy, freight, labor cost, to name a few. We reduced fixed costs in the plant network in Europe, and we signed new or expanded existing 4PRO solution contracts in most of the regions.
Our China team is using 4PRO successfully now to gain market share for the first time in several years and protect our price levels at the same time in this super competitive market. They are successfully developing, with their local partners, a cutting-edge robotic solution for ladles that we are excited to roll out very soon, first in China and then globally. Also, the strategic cooperation with robotics and automation specialist Polytec, signed in October last year, starts getting to life. The first customer project of this partnership is already close to full operation. An encouraging pipeline has been developed with the very competent Polytec team, which also fits very well to RHI Magnesita culture, actually, from the people side. Steel production around the world was soft amid ongoing volatility in the first half.
World Steel Association data through June 2026 showed a slight increase in crude steel production year-to-year. When removing the countries contributing most to the decline, China, Russia, Iran, Ukraine, and a few more, the remaining regions look more constructive, although starting from a very subdued level. Growth rates are difficult to take as naked numbers because the base is so low. We see the steel business on a stronger trajectory than previously because steel demand recovers locally. It's also going to be supported by trade barriers eventually and Chinese steel export coming down, although only very slowly and still staying at almost record levels. We also face challenges in the steel business. We lost temporarily some market share in the U.S. and in India.
In India, we consciously focused on sustainable growth and exited low-margin business, and we are focusing much more on 4PRO-enabled growth. In the U.S., the ERP go live temporarily impacted customer service levels and our market shares negatively in the second quarter. Both of those are short-term effects, and already starting to recover. Additionally, our over proportionate exposure to the GCC countries, to the Gulf countries in the Middle East, Turkey, and Africa region impacted us more than world steel production data suggests. Conversely, shipment growth in China demonstrates we can make the difference with 4PRO, as already elaborated, even in super strong commoditized market. Let's go to our industrial business. We need to look at our industrial business from two perspectives. The cement business inside this industrial division performed well. 2026 had an unusually weak start for our industrial projects business, however.
Overall revenues declined even more, namely by 13%, one three, to EUR 466 million as a result of a decrease in the industrial projects business. That is from a historically low 2025 level already. The overall numbers of industrial projects did not meaningfully recover. Fewer projects resulted in lower volumes, weaker product mix, and fixed cost under-absorption in the key high-cost plants, mostly in Europe, but also in a few other regions. These are very complex plants that make these products for these projects, and they are very expensive and they have been very underutilized. Our glass business especially was close to the bottom. Industrial applications, other industries investments was similarly weak. The non-ferrous project business was more resilient, with earnings in the first half performing at the level of our expectations, at least. Customers in all markets, in all regions exercise caution around CapEx decisions.
That is what drives the industrial projects. As a result, about EUR 50 million of high margin project revenue has been deferred already into 2027 from the first half of this year, and another EUR 10 million moved into the second half of this year. Only a small number of projects were totally canceled, most of those in the Middle East. Looking forward, we expect an improvement in the industrial projects business in the second half based on the order book. These stronger industrial sales are supported by the Northern Hemisphere cement maintenance season in the fourth quarter, that we expect like we see it every year. The order books of non-ferrous and glass have started to recover but they remain well below historic norms. We are breathing a little bit easier, but not relieved.
This will support fixed cost absorption in high key cost plants a little bit more and give an improved product mix, will also result in meaningful inventory reductions in the second half of this year. A couple of thoughts around the development of our strategy, ladies and gentlemen. To put the results in a bit more longer-term context. The industrial projects, which of course are the concern now, have strong fundamentals. This business remains a cornerstone of RHI Magnesita's competitive offering and also of our earnings performance today and especially in the future. While the project number has been much lower than expected, we believe this is a temporary effect and that the long-term fundamentals remain strong. These projects represent the most demanding refractory applications. Furnaces can operate up to 20 years. Processing high-value products under strict safety, reliability, and efficiency requirements is key.
A single project typically involves hundreds of different refractory products that together in a complex installation and in a certification process has to be delivered. Only a handful of suppliers can deliver on time and on quality all these products and services at the same time. The refractory material costs represent a smaller share of the total project investments of our customers. A failure here carries significant operational and financial consequences for our customers. Therefore, they continue to rely on RHI Magnesita's market leadership and the decades of experience. New 4PRO contract elements are under development and are being introduced also into these complex projects as they become available to our customers. We expect the industrial project demand to normalize slowly over the next 12 - 24 months with a stronger second half and a solidly looking 2027 from today's perspective.
The cement and lime business is our largest industrial end market, actually. Although projects represent a small share compared with the annual maintenance business. Growth opportunities are concentrated in developing markets, of course, where capacity is expanding to meet local cement demand, and in mature markets where kiln upgrades improve energy and efficiency and environmental performance. We are mitigating the project weakness quickly through closer OEM partnerships, the companies who build these cement plants, and the rollout of our 4PRO maintenance model, which is finding, I would say, enthusiastic reception from our customers. Recent innovations, particularly laser scanning technology for kiln wear monitoring, connected with AI-based optimization and attractive software visualization, are transforming maintenance practices and seeing rapid customer adoption pushed by customers' management teams. The non-ferrous metals offers in the non-ferrous metals business, we see the strongest long-term growth potential.
This is supported by strong demand for metals such as copper, nickel, platinum, and for lithium, all required for the electrification for data centers and also for the long-term decarbonization. RHI Magnesita is the clear market leader. For example, we are supplying most copper smelters outside China in the world and inside China, we still have a significant market share. The deliveries in non-ferrous metals projects declined to 38 projects last year compared with a historic range of 45 - 55. We believe that annual demand can increase to around 60 projects in the midterm. We are leveraging this expertise also in the green steel market, where new furnace designs require advanced refractory solutions at similar complexity than for the non-ferrous metals materials. Therewith, we have the ability to translate that knowledge. It results in several high-profile project wins that we have already gained.
The glass business remains our weakest larger end market. It is expected to stay below historical activity levels despite an improving order book that we see at the moment and that will deliver in the second half of 2026. We are addressing this through network optimization, reduction of capacities, and a sharper commercial strategy also. Finally, the industrial applications segment, all the other industries, span a very diverse range of technically demanding industries where our market share remains very low because we haven't focused on it very much in the past. Project deliveries often establish long-term customer relationships here. Meaningful growth is most likely through acquisitions of those kinds of portfolio.
The Resco acquisitions in North America has strengthened our position in that region, and the P-D acquisition in Europe has strengthened our position there, while other regions continue to offer attractive expansion opportunities based on the portfolio of those two transactions. Finally, before I turn over to Ian to dive into the details of the financials, let me have a few words on our network optimization, because this is one of the key drivers of profit improvement at the moment. The network optimization delivers sustainable savings and it drives our cash return on invested capital. We have built valuable experience in the last years to execute these plant mergers. Every acquisition that we make offers new optionality to optimize our network in a generally oversupplied industry. In all regions in the world, there's too much refractory capacity.
We are operating this and making these decisions in a very volatile market backdrop. We delivered the guided savings with our efforts in Europe. Spending on restructuring is behind us. The big spending and the guided savings are happening and they are reoccurring. However, we continue to work on our plant footprint in Europe as our European team uncovers new potential for cost optimization and customer service improvements at the same time. European plants focus more on local demand, but still optimize for export of sophisticated high-value refractory products. The first wave of self-help measures is advanced, and we have already started looking into the second wave in Europe.
In the Americas, the Resco acquisitions and the new trade environment created an opportunity to accelerate plant network adjustments in the U.S., especially, where customers are asking for more local-for-local supply and are pushing us for more secure supply chains, including security in raw materials. Onshoring is the main driver here, with the main benefit of shorter supply chains and for RHI Magnesita, lower invested inventory in the entire chain. We are pre-planning now measures to realize within the next 18-24 months and follow our customers' needs more aggressively. In Latin America, the opportunity lies more in restructuring, meaning concentration into larger flagship plants, notably Contagem in Brazil and Tlalnepantla in Mexico. This will increase the local-for-local share in both U.S. and Latin America and unlock fixed cost savings at the same time.
I am going to hand over to Ian and let him walk you through the financials in more details. Ian?
Thank you, Stefan, and good morning, ladies and gentlemen. As Stefan highlighted, I will walk you through our first half 2026 financial performance and then our expectations for the full year. The first half of 2026 demonstrates the strength and the resilience of our operating model. Adjusted EBITDA increased by 17% year on year from EUR 141 million to EUR 165 million, and our adjusted EBITDA margin improved from 8.3% to 10.3%. We faced a EUR 24 million foreign exchange headwind, mainly driven by the weaker U.S. dollar and the Indian rupee against the euro. On a constant currency basis, adjusted EBITDA increased by 42% year on year.
This improvement reflects the continued delivery of the group's pricing actions and structural cost self-help measures across our plant and SG&A in line with our expectations for this year. These benefits were partly offset by weaker demand for high margin industrial projects in glass and industrial applications. Despite the challenging environment, we defended our margins. Since the 2017 merger, our adjusted EBITDA margin has remained above 11% every year. This consistency reflects our diversification across regions and end markets, but more fundamentally, it reflects disciplined execution and active management. The refractory margin temporarily dipped below 10% in the first half, mainly due to weaker demand in high margin industrial projects. As Stefan highlighted, the resulting fixed cost under absorption in key high cost plants, particularly in Europe, could not be fully offset by self-help measures.
We are maintaining our full year refractory margin guidance at around 10.5%, supported by continued self-help, improving performance in steel, and the expected step up in industrial projects. The backward integration margin remains at its cyclical low, and we continue to expect only around one percentage point for the full year. We remain focused on our raw material self-help initiatives, including increasing sales of magnesite-based raw materials into non-refractory markets and reducing raw material production costs. Consistent with our normal working capital cycle, working capital increased in the first half as we built raw material inventory ahead of a stronger second half order book. Foreign exchange also added EUR 22 million to our working capital. Accounts receivable temporarily increased as well. In North America, receivables were up EUR 40 million due to delayed invoicing in the May month end following the new ERP go live.
We expect this to unwind in the second half and to support cash generation. The temporary inventory build and higher receivables will ease in the second half, and we are retaining our full year guidance of around 22% working capital intensity. As working capital increased in the first half, our net debt rose by EUR 33 million to EUR 1.528 billion. The group's leverage ratio remains stable at 2.9 times net debt to adjusted EBITDA. While operating cash flow was temporarily softer in the first half at EUR 160 million, cash generation remains a core strength of our business, and we again delivered strong cash conversion of 97%.
A stronger order book in the second half, continued delivery of self-help gives us flexibility to reduce net debt in the second half, and we therefore continue to expect net debt to reduce to approximately EUR 1.4 billion and leverage to move towards 2.6 times by the end of this year. We have also refinanced EUR 800 million of debt this year. This has increased our weighted average cost of borrowing slightly from 3.3% at the beginning of the year to 3.5% at the 30th of June, really only as a result of higher benchmark rates on the new debt, but our borrowing costs remain highly competitive. In line with our dividend policy finally, we've declared an interim dividend of $0.60 per share unchanged year-on-year. Finally, looking ahead, we are confirming our full year guidance.
We expect full year EBITDA to increase to EUR 435 million on a constant currency basis and to EUR 400 million on a reported basis, reflecting a EUR 35 million year-on-year foreign exchange headwind. The group remains on track to deliver self-help from pricing and cost measures, including the previously guided EUR 15 million EBITDA improvement from each of pricing and network optimization and the administrative cost savings. This self-help is being delivered in the current low demand environment, strengthening our group and improving operating leverage for when demand does recover. We are also pursuing further measures across raw materials and our plant network to sustain momentum in 2027 and beyond. We expect steel to contribute an additional EUR 25 million in the second half, driven by a stronger order book, particularly in India and META, and continued progress on 4PRO.
This step up was also evident, you will recall, in the second half of last year and has always been part of our full year guidance bridge. We expect industrial to contribute an additional EUR 35 million in the second half. The second half is typically stronger for the industrial segment. It's supported by the Northern Hemisphere cement maintenance season and by high margin industrial project sales. The industrial projects order book for the second half is showing growth across non-ferrous metals, glass, and some also in industrial applications in most geographies, albeit from a low base. As before, the risk remains customers delaying some of these high margin projects into 2027, particularly for geopolitical reasons.
In summary, we remain confident in stronger operational and financial performance going forward, supported by continued operational improvements, firmer order books in steel and industrial projects, and a clear focus on cash generation and de-leveraging.
Thank you. I'll now hand you back to Stefan for closing remarks.
Thank you, Ian. Just to summarize the first half again, the execution of our self-help measures delivers consistently. This was the driver of our improved performance in the first half. We will continue on this avenue and are in the process of identifying additional measures that will also continue this self-help supported improvement in 2027 and beyond. Second message. Our steel business, our cement business, and our non-ferrous business performed well, whereas the global industrial projects business remains challenging. We continue to see positive trajectory in steel. It's driven mostly via the rollout and penetration of our 4PRO business development. The industrial projects order book is rebounding, but from a very low level, and remains below historic norms. Third message. We confirm the full year guidance in a remaining very volatile market environment and rely on self-help measures more than on anything else.
Thank you very much for dialing in this morning. We're, of course, very happy now to listen to your questions and answer them as best as possible.
We will now begin the question-and-answer session. If you'd like to ask a question, please press star followed by one on your telephone keypad now. If you change your mind, press star followed by two. When preparing to ask your question, please ensure that your device is unmuted locally. You can also submit a text question via the webcast page. We will make a quick pause here for the questions to be registered. Our first question comes from Jonathan Hurn with Barclays.
Hey, guys. Good morning. Thank you for the presentation. I actually have three questions, if I may. The first one was to Stefan on some of the comments that you made in terms of your opening remarks. You're talking about the actions that you're taking now and the actions you're going to take going forward are going to improve the operating leverage of this business. I wondered if you could frame that a little bit better. Where do you think that the operational leverage in the business will go to on the back of these measures? Could we see drop-through at a much higher rate going forward than we have done historically? That was the first question. I can give you all three at the same time if you want. Okay.
Sure.
The second question was relating to a chart on slide five. It was in terms of Europe. Looking at that chart there, steel production was down. In terms of your volumes, you underperformed that, and it looks like you had some pricing pressure there. Can you fill us in a little bit more in terms of what you're seeing in that refractory market in Europe? The third and final question was on industrial projects. You've talked about that picking up, but can you talk about maybe the size of those industrial projects? I know the number is down year-on-year, but were you seeing any sort of difference in the value of those projects? Are they going down, or are they flat, or maybe they're increasing?
Those are the three questions, please. Thank you.
Let me ask Gustavo to answer the industrial projects outlook and pipeline because he's got really the best detailed view on this, and let me answer the other two questions first. Look, the operating leverage pull-through is about a 25% drop-through. Whatever additional growth comes will have a much superior profit contribution, and we would estimate that's the blended drop-through. It's higher in industrial, and it's a little bit lower in steel because of the nature of these plants. That's what we're expecting, and that is what we have calculated. Of course, with every measure that we take, we want that drop-through to increase, whereas these measures that we take are designed to reduce current costs and improve current performance, not just focus on the drop-through, because otherwise, we can't have a proper payback of any measure. Right? This is the focus here. Second question on Europe.
The European refractory market has always been split in two segments. There is the commodity brick per ton or mix per ton segment. That is a pure material specification delivered to the customer's plant, dumped on their doorstep, if you don't mind me saying it that way. That market already in the last 15 years was increasingly dominated by cheap Chinese imports through more or less traders, people who had stuff produced there and just shipped it to European customers. When markets were calm, that segment grew because it's less risky. When markets get more volatile, that segment shrinks because it's too risky for customers to rely on this. This is in Europe, I would say, about 1/3 of the refractory market.
The other two thirds are much more customized offerings in all market segments, from iron making to the casting line, but also in any kind of industrial application, non-ferrous, copper, cement, glass, and so on. What is happening here, the customers want a total service packaging, and that market is actually maturing more and more. With respect to RHI Magnesita, we managed to convert successfully some of this very commoditized approach at our customers to a more full solution approach. We can give you specific examples. I think we don't have time now to do this. This is the direction of the European market. From a volume perspective, total demand, we don't expect Europe to grow. Where should it come from? Construction is depressed. Transportation market is not growing very much. The machinery market is not booming.
The end market demand isn't pulling through the materials of our customers. Therefore, we will not see volume growth in Europe for the long term. Yeah, in the short term, from one quarter to the next, that might be different. Structurally, there's no growth in Europe. We need to be able to have a value improvement, and that comes from integrating our offering with our customers. Gustavo, would you mind to give a bit of a specificity on the industrial projects pipeline?
Sure. Firstly, let's talk about the size of these projects before talking about the pipeline. Of course, I think that here it's interesting to understand the impact that it brings to us. These projects, they can bring to us a revenue of about EUR 1 million-EUR 2 million or even EUR 10 million-EUR 15 million, depending on the size of it. That's the range that it fluctuates. Of course, it highlights the consequence that it brings for us. Obviously, the impact in our fixed cost absorption in our plants is also sizable when we don't have such a strong pipeline. Anyway, we need to be ready when it comes. That's the challenge that we face in this segment. From the customer perspective, we remain very strong with our offering. Our win-loss ratio is above 50% in most of the industries that we serve.
In ferrous metals, it reaches 70% win-loss ratio. Pretty much we win most of the projects that are available in the market. What we track is, of course, on the sales pipeline, how solid, how firm are our customers in the projected demand. During this year, we have some shipments that were postponed from first half to the second half, or even already to 2027. In the next rounds of interactions, for sure, we'll keep updating you about the development of this segment. Once again, we remain very strong with our offering. Our market share remains very solid. We continue to win the available products in the market.
The reason why we give you a number of projects is because there's enough projects that the average of the smaller ones and the bigger ones balance each other out.
Yeah.
The number of projects really is a good KPI because that shows the activity of our customers' industries.
Our market share development.
Yeah, our market development. Next question.
Perfect. Very clear. Thank you.
As a reminder, a star one on your telephone keypad to ask a question. Our next question comes from Jamie Murray with Bank of America.
Hey, guys. Thank you for taking my question. It is Jamie Murray from Bank of America. I had a couple of questions. The first one just about industrial projects, which clearly is the main source of uncertainty. I just had a couple of questions around the EUR 35 million industrial demand recovery that you expect in H2. I am just trying to gauge the level of confidence you have over that number. Could I please ask what proportion of that improvement depends on projects that have already been delayed at least once? Secondly, could I just ask about how much visibility you get before customers tell you about delays? Again, I am just trying to gauge how much additional slippage you could absorb before the EUR 400 million target comes under pressure and when you might know about it.
Separately, could I just hear about your latest views on how you see the price of magnesite and dolomite evolving over the next 12 - 24 months, and how that will impact your vertical integration margins? Thank you.
Thanks for the question, Jamie. The number, the EUR 35 million results from specific concrete order book entries. More than half of this, not two-thirds, but more than half has already started production. We are super confident on those because even if customers in those projects eventually come to the conclusion they want to delay their own CapEx project by a few months, usually we manage to deliver these projects. Customers take that because they know that, of course, we have already pre-approved, so we are very safe on those. The other ones are signed, dated, in full engineering, rollout, but with a bit more of a risk. That risk, of course, goes two ways. We can add a couple more projects, then the EUR 35 becomes EUR 40. We can also experience, usually this happens in November.
Around November timeline, when customers look at their CapEx budgets and they see they come under pressure and they get instructions from their top management to save on cash flow, then they move the CapEx into the next year. That happens every year, this is the risk that we run here, and I think we have to quantify it about this way. It is about a 60/40 super safe and 40 is more variable. Still, we remain confident because, of course, we also have a pipeline of projects that is not yet that firm that can move into this. It is business as usual, as happens every year, so it does not make us very nervous.
Jamie, just to add to that. Going into the year, like for steel, we expected a higher second half weighting on industrial demand, in particular because of the cement season, but also because we have a higher weighting traditionally on industrial projects. That was 20 of that 35.
Yeah.
15, you will recall on the full year bridge, was the improvement year-on-year. Indeed, part of that we did expect to come through in the first half of the year. That didn't happen, and it's now fully weighted to the second half.
Yeah, super. Thank you very much, Ian. On the magnesite price, there's a modest increase of fused magnesia and dead-burned magnesia prices, but modest. This is mostly driven by the fact that in China, the caustic kilns that have been operating under very dirty environmental conditions have been stopped in operations, almost all of them. There's a very good recovery on the caustic magnesia, which is the precursor product for the ones that we use. We benefit a little bit from this. We will in the second half. That's why our backward integration margin is a bit higher in the second half than in the first half because we have a good order book for those types of products. The pricing on the other materials is moderately increasing. I think we remain skeptical that this is a structural improvement at this point in time.
Capacity is reducing in China in the sector, we are nervous that this shortage on the caustic is contributing to this short-term price increase and it will live itself out of the system again next year once these old kilns have been replaced. That's a bit the nervousness. That's why we're not so bullish yet on the backward integration.
Perfect. Thank you.
Thank you very much. Next question, please.
Our next question comes from Harry Phillips. The scope for additional cost reduction, particularly in local for local in the U.S., and well developed is the transition to the global service model.
Harry, morning. Thank you for the question. As you've seen, we have delivered fully on our network optimization program in Europe. We guided that we would do EUR 10 million in 2025, EUR 20 million in 2026, and EUR 30 million in 2027. We are delivering well against that and have high confidence in that. As we've highlighted, there are further network optimization opportunities both in the Americas and in Europe. In the Americas in particular, we would expect savings in double-digit millions, but we have not yet guided on those figures. Certainly, in aggregate, we expect a lower capital intensity to be able to realize those savings.
The latter, Harry, is from a cash return on invested capital, actually the interesting piece. That's one time step down in supply chain, the capital in the supply chain. That in the U.S. is very attractive because this is where our longer supply chain is. As Ian mentioned, we are working through these numbers. Hopefully in November, we can give you a bit more of a clear guidance on it.
Thank you. Just as a reminder, dial star one on your telephone keypad to ask a question, or you can also submit a text question via the webcast page.
There are no more questions, I take it?
We have no further questions in the queue, I'll hand back over to the team for any final comments.
Okay. Thank you very much, ladies and gentlemen. Thanks for dialing in this morning. I know it's a big day for you all. We will leave you now. Thanks for your interest. We look forward to interacting with you over the course of the next days and weeks. See you after the summer. Goodbye from Vienna.
Goodbye.
Thank you, everyone. This concludes today's call. You may now disconnect. Have a great rest of your day.