Ferroglobe PLC (GSM) Q4 FY2025 Earnings Call Transcript & Summary

Good morning, ladies and gentlemen, and welcome to Ferroglobe's Fourth Quarter and Full Year 2025 Earnings Call. [Operator Instructions] As a reminder, this conference call may be recorded. I would now like to turn the call over to Alex Rotonen, Ferroglobe's Vice President of Investor Relations. You may begin.

Good morning, everyone, and thank you for joining Ferroglobe's Fourth Quarter and Full Year 2025 Conference Call. Joining me today are Marco Levi, our Chief Executive Officer; and Beatriz Garcia-Cos, our Chief Financial Officer. Before we get started with some prepared remarks, I'm going to read a brief statement. Please turn to Slide 2 at this time. Statements made by management during this conference call that are forward-looking are based on current expectations. Factors that could cause actual results to differ materially from these forward-looking statements can be found on Ferroglobe's most recent SEC filings and the exhibits to those filings, which are available on our website at ferroglobe.com. In addition, this discussion includes references to EBITDA, adjusted EBITDA, adjusted gross debt, adjusted net debt and adjusted diluted earnings per share, among other non-IFRS measures. Reconciliations of those non-IFRS measures may be found in our most recent SEC filings. We'll be participating in the BMO Metals, Mining and Critical Materials Conference in Hollywood, Florida on February 23 and 24. We hope to see you there. With that, I'll turn the call over to Marco.

Thank you, Alex, and thank you all for joining us today. We appreciate your continued interest in Ferroglobe. While 2025 presented significant external challenges, including muted demand, tariff uncertainty, delayed trade measures and elevated levels of predatory imports, it was a year in which Ferroglobe made important strategic progress and substantially strengthened its position for future growth. Most importantly, we achieved significant and impactful trade measures in both the European Union and the United States. In Europe, the European Commission voted to protect the ferroalloys industry by implementing safeguards, targeting a 25% reduction in imports relative to the baseline of average imports by country and product from 2022 through 2024. During those years, annual imports of ferrosilicon averaged approximately 450,000 tons and manganese base alloys averaged approximately 900,000 tons. These safeguards create a substantial opportunity for domestic producers, including Ferroglobe to regain market share under a more balanced competitive framework while ensuring security of EU supply chains for critical and strategic materials. We are encouraged by the European Commission's advocacy to support and strengthen the long-term sustainability of local industry. To further enhance the EU manufacturing base and drive economic growth, the Made in Europe pledge was signed by more than 1,000 business leaders. This is similar to the Buy American pledge, encouraging increased use of products with domestic content. In the United States, the International Trade Commission ruled in favor of imposing antidumping and countervailing duties on ferrosilicon imports from Brazil, Kazakhstan and Malaysia after having ruled similarly against Russia in 2024. These decisions meaningfully improve the long-term outlook for the U.S. ferrosilicon market. To capitalize on improving ferrosilicon economics, we have converted three furnaces from silicon metal to ferrosilicon, one in the U.S. and two in Europe. This highlights the benefits of our diversified global footprint, which enables us to optimize production in response to market dynamics and geopolitical factors. With respect to silicon metal in the U.S., the case was delayed due to the government shutdown. Prior to the shutdown, the preliminary decision in September indicated strong measures against Angola, Australia, Laos, Norway and Thailand. The preliminary combined antidumping and countervailing duties range from 21% for Norway to 334% for Laos. We now expect a final decision on Angola, Laos and Thailand later today with Australia and Norway anticipated in June. Operationally, we executed with discipline and focus through proactive cost control measures, including a hiring freeze and reduced discretionary and CapEx spending, we successfully navigated through weaker demand and lower pricing while maintaining a solid balance sheet. After the Safeguard announcement on November 18, CRU index prices for ferrosilicon and manganese alloys in Europe jumped approximately 20%. While ferrosilicon has retreated some in recent weeks, it is still up more than 10% since the Safeguard announcement. Our outlook for silicon metal remains more measured due to its exclusion from EU safeguards and continued aggressive imports from China and increasingly from Angola. In the U.S., the silicon market is expected to grow modestly according to CRU. We are actively studying longer-term opportunities associated with our idle operations in Venezuela. This site includes three large ferrosilicon furnaces and a manganese alloy furnace originally designed to produce silicon metal which can be converted back to silicon metal. In addition, the facility includes a Soderberg-based plant that could be used to produce electrodes. While it is too early to determine the timing and condition of the infrastructure and operations, the asset base represents a potential opportunity for the future. Given Venezuela's proximity to the U.S. market, this opportunity could become strategically meaningful over time. We also took important steps to enhance our long-term cost structure and operating flexibility, signing a new competitive 10-year French energy agreement effective January 1, 2026. In addition to competitive energy prices, this agreement provides greater flexibility, enabling us to produce up to 12 months a year in France. Combined with implementation of safeguards, this flexibility meaningfully improves the earnings potential of our ferroalloys business by allowing higher volumes to leverage our fixed operating costs. Beyond our core operations, we continued to invest in long-term opportunities, increasing our total investment in Coreshell to $10 million in 2025, reflecting strong technological progress in the development of advanced silicon-rich EV batteries. In addition to ongoing collaboration with automotive OEMs, Coreshell is expected to begin initial shipments to defense and robotics customers in the first quarter of this year. Furthermore, we are in the process of finalizing the multiyear supply agreement with Coreshell. For those who are new to the Coreshell story, silicon rich anodes offer lower cost batteries with increased capacity, longer driving range, faster charging and maybe most importantly, a reduced reliance of graphite, of which more than 90% is produced in China. We believe this technology has the potential to become increasingly strategic over time. Alongside these trade developments and operational enhancements, we continue to execute on shareholder-friendly capital allocation. We increased our first quarter 2025 dividend by 8% to $0.014 per share, and we are increasing it again by 7% to $0.015 per share starting in the first quarter of 2026. In addition, during the early part of 2025, we selectively executed discretionary share repurchases, acquiring 1.3 million shares at an average price of $3.55 per share. Looking forward to 2026, Ferroglobe is well positioned to benefit from the cumulative impact of the trade actions. We expect most of our segments to post considerable growth in 2026, and we anticipate revenues improving to a range of $1.5 billion to $1.7 billion, an increase of 20% at the midpoint over 2025. This expectation is driven primarily by strong volume growth in the silicon-based and manganese-based alloys segment. The implementation of EU ferroalloys safeguards and the U.S. ferrosilicon antidumping and anti-circumvention rulings give us increased confidence in this outlook. Next slide, please. Our shipments increased by 13% to 165,000 tons on the strength of silicon-based and manganese-based alloys, resulting in a 6% increase in quarterly revenue to $329 million. Our adjusted EBITDA declined slightly to $15 million, while our free cash flow was negative $19 million. Beatriz will provide more detailed comments in her section. Next slide, please. I'll update on our segments, starting with silicon metal on Slide 5. This may sound like a repeat of last quarter, but the situation remains essentially unchanged. The demand is still weak across our regions, and Europe is still plagued by unabated predatory imports from China, which roughly doubled in 2025 as well as by rising imports from Angola up nearly fourfold, driving prices to unsustainable levels. As a critical and strategic material, the European Commission should ensure sufficient EU production to meet basic demand. Overall, volumes and revenues declined by approximately 3% due to an 8% decline in U.S. shipments, partially offset by a 5% increase in shipments in the EU. It is important to note that the EU shipments in the fourth quarter were up from a very weak third quarter. We idled our EU silicon metal plants in the fourth quarter due to the extremely low unprofitable prices. Within the Silicon Metal segment, the aluminum sector is performing better in relative terms as highlighted by the recent recovery in aluminum prices compared to the weak polysilicon sector. The chemical sector remains weak also due to imported siloxane and silicones from China into Europe and the U.S. The U.S. index prices rose a modest 2% in the fourth quarter over the third quarter. EU prices declined by 7%, primarily due to imports. For the year, European prices are down by 1/3, while U.S. index is up less than 2%. In the U.S., we expect the volumes to improve in the second half of 2026 as the antidumping and anti-circumvention measures are expected to be finalized in February and June. Next slide, please. The story is quite different in our other product segments. Globally, silicon-based alloys had a very strong fourth quarter. Total volumes increased by 19% to 51,000 tons with EU and North America increasing 25% and 14%, respectively. Pricing trends were mixed in the fourth quarter. The EU ferrosilicon index rebounded strongly in the quarter, rising 22% to EUR 1,495 from Q3, driven by the implementation of safeguards in November. In the U.S., the ferrosilicon index retreated a modest 4% during the quarter. For the full year, the European index gained 12%. The U.S. index is down less than 2% for the year. Overall, we are optimistic that 2026 will be a stronger year for total silicon-based alloy sales for Ferroglobe. We have already booked incremental business for 2026 in Europe and in U.S. An additional catalyst for the second half of the year is expected from enhanced EU steel safeguards with a proposal to reduce import quotas by 50% and double tariffs to 50% for exceeding the quota. It is anticipated that domestic EU production will be ramped up as a result. These measures are expected to take place on July 1, 2026. Next slide, please. Our manganese segment reported another strong quarter with a 16% volume increase to 81,000 tons, up from 70,000 tons in the third quarter. We benefit from a larger customer base as well as safeguards. EU sales, which accounts for more than 90% of our manganese volumes grew 18%. Manganese alloy index prices improved substantially in the fourth quarter with ferromanganese and silicon manganese increasing 16% and 21%, respectively. The combination of solid demand from our European steel customers whose business is expected to grow by 3% in 2026 in EU and safeguards should propel a robust volume increase in 2026. Accordingly, we are optimistic about the European market opportunity for manganese this year. I would now like to turn the call over to Beatriz Garcia-Cos, our Chief Financial Officer, to review the financial results in more detail. Beatriz?

Thank you, Marco. Please turn to Slide 9 for a review of the fourth quarter income statement. Fourth quarter sales grew 6% over the prior quarter to $329 million, while raw material costs increased 23%, excluding the $40 million impact of [ PPAs ] for all power purchase agreements. Fourth quarter raw materials and energy as a percentage of sales increased from 58% to 67%, primarily due to temporary idling in France, again, excluding [ PPAs ], all purchase price agreements. These [ PPAs ] are mark-to-market using fair value. Given the long-term nature of our EDF contract, which accounts for the majority of the [ PPA ] impact, we will continue to strip out [ PPA ] mark-to-market volatility to provide a clearer view of our underlying operational performance. A strong silicon base alloys and manganese base alloys volumes drove the increased sales with quarter-over-quarter volumes increasing 19% and 16%, respectively. Silicon metal volumes declined by 3%. Volume improvements were partially offset by a 6% decline in average selling price of silicon-based alloys and manganese-based alloys with silicon metal prices essentially flat. Adjusted EBITDA declined 20% from the prior quarter to $15 million versus $18 million. Adjusted EBITDA margin declined to 4%, driven by lower prices and elevated costs as a result of idling in France. Next slide, please. Moving to product segment, adjusted EBITDA bridges. Silicon metal revenue declined 3% sequentially to $96 million in Q4, driven by a 3% decrease in shipments to 33,000 tons. The average selling price was essentially flat at $2,957 per ton. Volumes remain constrained due to soft markets in the U.S. and by Chinese and Angolan dumping of silicon metal in the European Union. Silicon metal adjusted EBITDA declined from $12 million to $1 million in Q4, with margins decreasing to 1%. The margin compression was primarily due to idling in France, slightly offset by improved cost in North America. Next slide, please. Silicon-based alloys revenue grew 12% to $104 million, driven by a 19% sequential increase in volumes to 51,000 tons, partially offset by a 6% decline in average selling prices to $2,020 per ton. Adjusted EBITDA increased to $16 million in the fourth quarter from $12 million in the third quarter. Margins expanded by 160 basis points to 15%. The improvement in adjusted EBITDA and margins result from lower cost in Spain, partially offset by early idling in France. Next slide, please. Manganese base alloys revenue increased 10% to $93 million from $84 million in the prior quarter. The improvement was primarily due to a 16% increase in volumes to 81,000 tons. Fourth quarter average selling price declined 6% to $1,147, mainly due to a lag versus index prices. Adjusted EBITDA in the fourth quarter doubled to $9 million, while adjusted EBITDA margins increased from 5% to 9%. This margin expansion was primarily driven by improved performance in Norway and higher overall volumes. Next slide, please. Adjusted EBITDA for the full year was $28 million, down from $154 million in 2024, and the adjusted EBITDA margin declined to 2%. The price decline driven by weak demand and increased imports to Europe had a significant impact on adjusted EBITDA, accounting for more than 80% or $104 million of the decline. Reduced volumes account for another 16% of the EBITDA decline. Cost impact on adjusted EBITDA was negligible, while head office and noncore business detract less than $3 million from adjusted EBITDA. Next slide, please. There are lots of details on this slide, so I will just highlight a few of the most important items. For the full year, we generated $51 million in cash from operations, driven by a $48 million improvement in net working capital. We curtailed CapEx by $6 million to $63 million in 2025. For the year, our free cash flow was negative $12 million. During the fourth quarter, we consumed $4 million in operating cash flow due to weak EBITDA and an increase in net working capital of $8 million. For the year, we reduced our net working capital by $48 million, in line with our target of $50 million. Energy rebate was $7 million for the fourth quarter. As we operate under the new contract in France in 2026, we don't expect energy rebates going forward, which will help align our adjusted EBITDA generation more closely with our cash flow. Fourth quarter capital expenditures totaled $14 million, representing a $5 million reduction versus the third quarter. Next slide, please. Despite the headwinds in 2025, our balance sheet remains strong. In total, we paid $10.5 million in dividends during the year, and we are again increasing our dividend. Starting in the first quarter of 2026, our quarterly dividend will increase 7% to $0.015 per share. It will be paid on March 30 to shareholders of record on March 23. While we did not repurchase any shares in the second half, we bought back 1.3 million shares for the first half. Our discretionary share repurchase plan remains in place. Our net debt position increased to $30 million in 2025, and we remain in a solid financial position to support growth in 2026. We reduced our 2025 CapEx by 20% to $63 million. At this time, I will turn the call back to Marco.

Thank you, Beatriz. Before opening the call to Q&A, I'd like to provide key takeaways from today's presentation on Slide 15. 2025 was a year of important progress for Ferroglobe. The trade measures secured in Europe and in the United States represent a clear positive shift in our markets, particularly in ferroalloys. Europe safeguards and U.S. antidumping and countervailing duty rulings significantly improved competitive conditions, support pricing and give us increased confidence in a much stronger market environment in 2026. At the same time, we continue to execute a disciplined shareholder-friendly capital allocation strategy. Despite the challenging macro backdrop, we increased our dividend during 2025, completed selective share repurchases and have announced another dividend increase beginning in the first quarter of 2026. These actions underscore our confidence in the business and our focus on delivering consistent shareholder returns. We have also taken important steps to enhance the economics and flexibility of our operations. The new long-term energy agreement in France, combined with our ability to shift production from silicon metal to ferrosilicon, allow us to optimize volumes, better leverage fixed costs and respond efficiently to changing market conditions across our global footprint. At the same time, we managed through a difficult demand and pricing environment in 2025 with discipline and focus. Proactive cost control, strong execution and a solid balance sheet allowed us to navigate near-term headwinds while strengthening the foundation for future growth. Overall, we believe Ferroglobe is exceptionally well positioned for 2026 and beyond. With improving market fundamentals, increased confidence driven by trade actions and more flexible, efficient operating platform, we see a clear path to stronger performance and long-term value creation for our shareholders. Operator, we are ready for questions.

[Operator Instructions] And the question comes from the line of Martin Englert from Seaport Research Partners.

I wanted to touch on volume expectations across the 3 businesses for 2026 and then also the plan for the EU silicon assets. I know you provided some detail about the furnaces converting over to ferrosilicon, but what remains idle there? And is it to remain idle for the foreseeable future, but any type of goalposts for volumes on silicon metal in 2026, silicon-based alloys and manganese-based alloys would be helpful.

Thank you, Martin. Let me try to address your first question on volumes, and then I move to the assets. Starting from Europe, the safeguards basically on ferrosilicon and ferrosilicon kind of products free up 25% of imports that were 450,000 tons in 2025. So 25%, about 100,000, 110,000 tons of these products are made available for EU 27 producers. For manganese, the imports were around 1 million tons. So when you calculate 25% of it, you end up to 250,000 tons available for EU 27 producers. And of course, all this pie has to be shared among the local producers, assuming that safeguards are controlled and put in place in a proper way. When you go to the U.S., I think that we are at the end of the period where inventories, mainly of Russian product, but also other products have been waiting on volumes and also price recovery. So we expect some gains in ferrosilicon, and we see that already from our customer portfolio in U.S. in the first quarter 2026. On top, talking about trends, steel will improve -- is expected to improve in Europe by about 3% across the year, mainly in the second part of the year when the new safeguard measures imposed by EU will be applied with a further reduction of imports of 50% and increase of tariffs to 50% for excess of products. Aluminum is expected to grow in Europe by a solid 3% next year. In U.S., aluminum is expected to grow between 8% and 9%, while steel is expected to start recovering. And actually, we have seen asset utilization in U.S. recovering already in quarter 4 2025. These are the major indicators for volumes in ferroalloys, both in U.S. and Europe. Going to your second question, yes, we have converted one furnace in Beverly to ferrosilicon already during 2025. We have converted as of January two furnaces in Europe, one in Salon and one in Loudun from silicon metal to ferrosilicon. And concerning the utilization of the other silicon metal furnaces in U.S. and Canada, we're fully utilized. While in Europe, we are selectively restarting furnaces based on contracted demand. So some furnaces will stay idled in the first quarter, while some others have been restarted to supply contracted volumes already in January. Sorry for my voice.

I appreciate all the detail and context there. I wanted to inquire about the component of minimum prices with EU safeguards for ferroalloys. Do you ultimately expect that domestic prices will gravitate to these levels? Or is there a dynamic within the EU footprint where there's sufficient capacity out there and that there isn't necessarily maybe the case that we see parity with the minimum price levels embedded in the safeguards?

Yes. Well, the key question is demand, which is not until now has not been great or has been muted both in Europe and in U.S. So the key question is how much demand is going to ramp up for the different products. There is definitely enough capacity in EU to cover the safeguards for all products. The -- if you look at what happened until now in Europe in ferrosilicon prices have jumped up by 22% on ferrosilicon immediately after safeguards have been announced. But then due to stocks at traders and others, the price -- index price has been going back. And today is only 10% higher than previous safeguard announcement. Different trend for manganese. Manganese products have jumped up around 20% on average in terms of index price. The price is holding. It's not improving, but it is holding at this level. And this is why I say that demand is critical. And again, I expect a major improvement of steel demand in Europe in the second half of 2026.

Do you know if Ukraine's manganese alloys facilities are still producing and supplying just coming into the region overall?

Yes, they are, but at a very small rate for reasons that are related to supply chain, but also conditions of the assets. Of course, I do not have too many detailed insights about the status of the assets, but considering the number of years of war, the location of the assets, I think that even when this is signed, if safeguard gets signed, it will take a while before they ramp up to the previous rates.

I understand. Are you able to explain a little bit and provide some context about how the EU carbon credits function, what's covered with your allocated carbon credits for 2026 volumes? And maybe discuss if you have to go back to the carbon credit market for incremental volume output that you may gain from market shares due to safeguards?

So this is a question that requires about one day of explanations, but let me try to be short. First of all, at the moment, on CBAM, we are impacted only for high carbon ferromanganese, not for the other products. And the way that the CBAM works basically looks at the imported products, looked at the content of actually the emissions of CO2 per ton required to produce these products, and they apply to the amount, the cost of CO2 in Europe per ton, deducting whatever the supplier has paid in his own country for its CO2 emissions. So the overall game is to tax CO2 heavy exporters to Europe to favor European producers who are at lower -- producing with lower CO2 emissions. Now all of this would be beautiful, was beautiful. If, one, there was a proper calculation of the CO2 emissions for every kind of producer in the exporting countries. All of this will be beautiful if Europe was not anxious to reduce our CO2 credits because trying to implement these measures when you don't have all the data, you end up potentially penalizing more European producers than the exporters. So the -- of course, the commission is aware of that. They are doing their best. So steel is heavily involved in that. We will see how the situation develops. But again, we are -- our impact at the moment is minor due to the fact that CBAM is applied on the [indiscernible] carbon ferromanganese.

I appreciate all the color and detail and good job on the cost performance given the fundamental volume headwinds.

We will take our next question. The question comes from Nick Giles from B. Riley Securities.

My first question was maybe just back to silicon metal exclusion from EU safeguards. And just wanted to get your perspective on what the EU's appetite might be to revisit an inclusion of silicon metal in those safeguards. And maybe any background you can provide on kind of what prevented them from being included in the first place?

Yes. Well, the -- as you know, we asked for safeguards for silicon metal in Europe. The reasons why -- the official reasons why Europe did not support us on this request are related to the fact that silicon metal has a much stronger energy footprint, meaning it requires much more energy to be produced versus the other products. The other key element from a WTO perspective is related to the fact that imports did not increase in absolute terms. They have increased in relative terms for the period considered because the imports gained an 85% market share in the EU 27 territory. But in absolute terms, they did not. These were the main two reasons. The third reason was about our point is related to both that silicon metal and ferrosilicon are interchangeable which is hard to dispute because we can convert our furnaces from silicon metal and ferrosilicon and vice versa. And our customers in steel can move from ferrosilicon to silicon metal. So the fact that they called for non-interchangeability was quite a surprise due to the time and the number of meetings that we spent to explain our business. The top reason was a strong opposition of the chemical industry to protect silicon metal and a strong opposition of Germany to trade measures. And so there is a combination of -- there was a combination of technical, if you want, and political and legal aspects that excluded silicon metal from the safeguards. This situation doesn't make any sense for two reasons. One is that without protecting silicon metal, you basically don't protect ferrosilicon and because it's quite easy for steel producers to replace ferrosilicon with silicon metal when the price difference is not there. Silicon metal should be much more expensive because of the energy, which is required to produce it. So approving safeguards for ferrosilicon and then not for protecting silicon metal is like shooting in your own foot. And this shouldn't be a surprise for the European community. Going to the first part of your question, yes, we are working on new measures for silicon metal in Europe. Actually, the commission has asked us to submit our data. So we have submitted them mid-December relative to the last year in Europe. And we are waiting for their reactions to this document. The expectation is that we will go for antidumping to -- against the major dumpers in Europe. China is number one. And then we are still to see how do we address Angola, which is a sort of subsidiary of China with the same pricing policy. The combined volume of the two countries, if you calculate also Vietnam, which is pure circumvention of silicon metal from China, basically shows that last year, the volume coming from China or China subsidies doubled. And in a market that has been going down, of course, with pure liquidity, they have been determining price based on index. And like I reported in the last quarter, they go to market with prices 25%, 30% before -- below the cash cost of European producers. So it's clear dumping. The case is clear. The difficulty is not the case, is the fact that, as you know, Europe and the industry in Europe is finally reacting to these situations. And so the commission is -- the commission table is jammed with cases asking for protection for every kind of products. And this causes some delays. So at the moment, the game is all about politically pushing to get our case at the top of the pile of cases of the European Commission.

Marco, I really appreciate all that background and your perspective on the situation. I guess my follow-up question to that is ultimately, there's plenty of reason for optimism in ferrosilicon when you look at that segment in a vacuum. But do you think that these dynamics within silicon metal could ultimately weigh on pricing volume expectations in FeSi?

Well, it all depends on -- yes, yes, it all depends on demand and appetite of steel makers to embark reformulating with silicon metal knowing that medium term, it doesn't make sense to have silicon metal at the price of ferrosilicon in Europe. At this stage, our order portfolio has started going up already in quarter 4 on ferrosilicon, both in U.S. and in Europe.

Understood. My next question would just be over the past couple of years and especially with the change in the administration in the U.S., I think end market exposure has shifted. And then you kind of layer on the conversion of some of your silicon metal capacity to FeSi that kind of reweights things more towards steel market. So I was just hoping you could kind of zoom out and provide us a high-level breakdown of your ultimate end market exposure. I think about solar as an area that comes to mind that might be less relevant today than it was in the past. So I appreciate any color you can provide there.

Well, today, when you look at our total business, I would say the 70%, 80% of the business is protected and only 20% is not, which is basically silicon metal in Europe. So the other high-level thing, it's not a surprise that the United States, apart from government shutdown are pretty favorable to protect critical and strategic minerals. And this is why we are going for antidumping, anti-circumvention and things are going fast. But it's also true that things are changing in Europe, maybe not at the speed that we would like. But in terms of political support, I have to assure you that our case is at the top of the agenda of all the states that are involved in ferroalloys and silicon metal. So the Europe has decided to protect industry -- has decided to protect our industry like the chemical industry or other industries. The problem is that Europe is not united like United States. So there is quite an exchange of -- continuous exchange of responsibilities between the center, the EU commission and the single states. The last case was last week when Von Der Leyen basically told the states were complaining about how slow the EU is in deciding, basically pushing back decision to the states. And this delicate situation in EU is causing [ outcomes ] much slower. On the other side, when I talk to politicians in Spain, in Norway and in France, they are pretty aware of what we need to do, which is a combination of protection, right, energy price for the industry and make sure that when they think about products, they think about supply chains and not about single products.

Very good. Maybe just one quick one, if I could, for Beatriz. I want to commend you on really managing the cash balance during this -- during the trough here. I mean you still have a pretty healthy net cash position. So can you just touch on anything we should be focused on from a working capital perspective? CapEx was down year-on-year. Should we kind of expect CapEx to be more flat this year? Anything I might have left out just from a cash flow or capital allocation perspective?

Yes. Thank you for the question, Nick. As you have been seeing on the deck, so the cash position at the end of the year is -- we finished the year with a strong cash position. Nevertheless, I have to say that this difficult year, the cash is coming mainly from the release of the working capital. So we have been releasing at the end, $48 million of working capital and therefore, a total positive operating cash flow for the year, right? Looking into 2026, I think one of the things that we are working and we are already seeing the results is the -- will be additional release of the working capital, even if the business we are planning to produce more volumes and sell a higher number of tons that this is typically creates consumption of working capital because we have this S&OP in place. We plan to continue to release additional working capital. And I refer to the target that we put out there, I think, in 2024, when we said that we want to run the company with a 20% less of working capital, if you remember. Now we are close to the $400 million, and we expect to continue to go down into the release of working capital. So that's one angle. On the other side, we are in a net debt position at the end of the year and we expect to improve slightly maybe as well this net debt position as we go during the year and of course, supported by the release of the working capital and cost reductions as Marco has been mentioning.

And then just CapEx, you would expect CapEx to be pretty similar to 2025 levels?

Yes. So this year, we went to $63 million for CapEx. That is already a 20% less versus 2024. And in 2026, we expect similar or slightly lower levels of CapEx. Of course, this is -- I'm always talking about maintenance CapEx rise of the company or sustaining CapEx, yes, maybe around the same number.

This concludes today's question-and-answer session. I'll now hand back for closing remarks.

Thank you, Heidi. We are encouraged by our accomplishments in positioning the company for a more robust market environment and much stronger financial performance in 2026. Thank you again for your participation. We look forward to updating you on the next call in May. Have a great day.

This concludes today's conference call. Thank you for participating. You may now disconnect.