Umicore SA (UMI) Earnings Call Transcript & Summary

Ladies and gentlemen, thank you for standing by and welcome to the Umicore Full Year 2019 Results Conference Call. [Operator Instructions] I must advise you that this conference is being recorded today, Friday, the 7th of February, 2020. And I'd now like to turn the conference over to your speakers today, CEO, Marc Grynberg; and CFO, Filip Platteeuw. Please go ahead.

Thank you, Joanne. Good morning, everyone, and welcome to the presentation of Umicore's performance for 2019. I will first make some comments on our performance and outlook, give an overview of our major achievements and then hand over to Filip, who can talk you through the 2019 financials. I will then wrap up before handing the call over to you for any questions that you may have. Despite the challenging market environment in 2019, particularly in the automotive sector, I'm proud to announce that we turned in a strong performance while making significant strides in the execution of our growth strategy in all 3 major activities. In Catalysis, we outperformed the car market by significant mileage as a result of market share gains in gasoline catalyst applications. This outperformance was most pronounced in China, the largest car market in the world, where we became the leading light-duty catalyst provider. We also recorded strong growth in the demand for our fuel cell catalysts. In Energy & Surface Technologies, our sales of cathode materials for electric vehicles grew in line with the market, although the 2 halves of the year were sharply contrasted. Sales were softer than the market in the first half and improved somewhat in the second half when the rest of the industry was down following the subsidy cuts in China. Results of the business group were affected by lower cobalt prices and the competition from unethically sourced cheap cobalt from artisanal mining as well as the higher depreciation charges and upfront costs related to our greenfield investments. In Recycling, we posted a very strong performance. While we processed lower volumes in Hoboken due to the scheduled extended shutdown in the first half of the year and the fire incident in July, we optimized in the feed mix in order to offset the effect of the volume shortfall. The business group also benefited from favorable supply terms and higher precious metal prices. With our statement in April 2019, we were among the first players to identify that demand patterns for cathode materials were deteriorating in the Chinese EV market, in particular. We were also amongst the first to project no near-term recovery in automotive demand in general. The statements we made and the outlook we gave at the time have, unfortunately, proven to be accurate. As we communicated then, we expect to grow revenues and earnings in 2020 despite a deterioration in the macroeconomic environment globally, in particular, in the automotive market. It is also worth to point out that visibility is limited today. The outbreak of the coronavirus is probably going to amplify the economic slowdown although it is much too early to estimate how much or for how long. Our growth outlook assumes today no protracted or material effect of the virus on economic activity in 2020. In Catalysis, although market projections point to slightly declining or flat car production at best, we expect to benefit from our strong market position in gasoline catalyst applications and a higher penetration rate of gasoline particulate filters in Europe and China. Fuel cell catalyst production will ramp up in our new plant in Korea and contribute to the growth. For Energy & Surface Technologies, we expect higher sales of cathode materials for EVs, and a positive contribution from the acquisition of the Kokkola plant. We do not expect EV sales in China to show a material degree of recovery. Our performance in 2020 will be affected by continued investments in greenfield production sites, resulting in higher depreciation charges and substantially higher start-up costs as well as increased R&D costs as we step up product and process development programs. In Recycling, the Hoboken smelter is expected to offer more availability in 2020 after completion of the expansion investments in 2019. We expect the supply environment to remain favorable and that metal prices, some of which were hedged in the course of 2019, will continue to provide tailwinds. Overall, I'm pleased to confirm a growth outlook for 2020, despite the adverse market trends that developed in 2019 and continue to prevail today. We will continue to execute our growth strategy with determination while keeping the agility to adjust our investment programs to take account of evolving market needs in the short term. Let's now turn to the review of each business group, starting with Catalysis. There was a significant contraction in global car production in 2019, actually, the strongest decline in car production since the 2008 recession. The biggest reduction was observed in the world's largest car market, China, which contracted for the second consecutive year. Diesel car production continued to decline in Europe and now accounts for a 35% share of the European car market. On a positive note, more stringent vehicle emission standards have come into effect in key regions and the market share of gasoline particulate filters increased in Europe and China, as expected. Umicore strongly outperformed automotive production, with revenues up by 7% year-on-year as a result of market share gains in light-duty gasoline and the growing penetration of gasoline particulate filters in China and Europe. We have now become the leading supplier of light-duty catalysts in China, reflecting market share gains and a strong exposure to platforms, which have implemented China 6 norms ahead of the due date. Precious Metals Chemistry also increased revenues year-on-year, with growing sales in pharmaceutical and fine chemical applications as well as in fuel cell catalysts. Throughout 2019, we continued to invest in R&D, supporting the transition to more demanding emission norms. Also, we expanded production capacity in China, Poland and India to cater for the growing demand for our catalysts. Very recently, we opened a new plant in Korea for the production of catalysts for fuel cells. Hydrogen fuel cells are gaining momentum as a clean mobility solution for light and heavy-duty applications. And Umicore is well placed to benefit from a growing market penetration of this technology. Let's now move to Energy & Surface Technologies where the market context also became more challenging last year. The global EV market grew by less than 8% in 2019 compared to more than 60% in 2018. It is also worth pointing out that 2019 was a year of 2 contrasting halves in the EV market. While global EV sales continued to grow in the first half, albeit at a slower pace than in 2018, EV sales decreased in the second half, as expected, due to an abrupt decline in EV sales in China following the subsidy cuts. In consumer electronics, there was also a slowdown in demand, which was caused by destocking across the value chain. In the energy storage market in Korea, demand was very much down owing to safety incidents on certain installations. Finally, cobalt prices were depressed to less than half 2018 levels, and this was exacerbated by the inflow of cheap cobalt products originating from unethical sourcing. Against the backlog -- against the backdrop of declining EV demand in the second half, we managed to grow our sales of NMC cathode materials, both sequentially and compared to the same period in 2018, as expected. For the full year, Umicore's sales of NMC cathode materials for EVs grew in line with the EV market globally, or, in other words, our market share has remained stable in this segment. As also expected, sales of LCO cathode materials for consumer electronics and NMC for energy storage applications were lower year-on-year, and we have no indications of an imminent recovery of demand in these segments. Financial performance of the Rechargeable Battery Materials business unit was also affected by higher depreciation charges and upfront costs related to our expansion investments in China and Poland as well as higher R&D costs. The Cobalt & Specialty Materials business unit was severely impacted by the lower cobalt price and low demand for end products as customers in several industries reduced excess inventories. We also continued to face unfair competition from cheap unethically sourced cobalt supplies, which have dented our sales volumes and margins for high cobalt-containing products. You may recall that we indicated last year that the combined effect of lower cobalt price and the unfair competition from unethically sourced cobalt was estimated at EUR 10 million to EUR 15 million in the first half of last year compared to the first half of 2018. For the full year 2019, we estimate this impacted approximately EUR 25 million. While we were adjusting to short-term fluctuations in demand with the required agility, we made significant strides in the execution of our strategy, which should enable us to capture significant future growth. On the supply side, we concluded long-term supply partnerships for sustainable cobalt with Glencore; and in January 2020, with CMOC. We completed the acquisition of the Kokkola cobalt refinery and precursor facility, which, together with the supply partnerships I've just mentioned, strengthen our sustainable value chain. We started commissioning of the greenfield plant in China and construction of the new plant for cathode materials in Poland. Downstream, we signed significant multiyear sales agreements with leading EV battery producers, LG Chem and Samsung SDI, for deliveries from our plants in Korea, China and Europe. Another milestone was the qualification for financial support for certain of our innovation programs under the umbrella of the Important Projects of Common European Interest, also known as IPCEIs. This umbrella was established by the European Union to provide a framework under which member states are authorized to provide financial support to projects which aim at creating a sustainable and innovative battery value chain for EVs in Europe. Finally, I'm pleased to report that the Global Battery Alliance, of which Umicore is a founding member, has now issued clear sustainability principles for the rechargeable battery industry. As the next step, the alliance will develop a battery passport which will trace the origins of materials and monitor them throughout the entire life cycle of batteries. This passport, which would act as a type of quality seal on global digital -- on the global digital platform, should help eradicate unacceptable practices from a social or environmental viewpoint and establish a level playing field. And obviously, we will seek widespread support from car OEMs for its implementation. In Recycling, 2019 saw a supportive environment in terms of metal price, notably, precious and platinum group metals, especially in the second half. Umicore also experienced a favorable supply environment with increasing availability of complex secondary materials such as spent automotive catalysts. And these tend to have higher metal loadings on average as the proportion in the mix of spent catalysts for tighter norms such as Euro 4 or Euro 5 and equivalent norms in other regions is growing. Also, China's Green Fence policy resulted in higher availability of end-of-life materials, such as printed circuit boards. Revenues of the Recycling business group increased by 9% in 2019 to EUR 681 million and EBIT by 40% to EUR 188 million, mainly due to the favorable supply mix and higher metal prices. In Hoboken, we successfully optimized the input mix in order to offset most of the shortfall caused by the extended scheduled maintenance in the first part of 2019 and the fire incident in July. Revenue at Jewelry & Industrial Metals remained stable year-on-year, while the earnings contribution from Precious Metals Management increased substantially due to favorable trading conditions, in particular, for PGMs. While the multiyear expansion program at the Hoboken plant was completed in 2019, investments continued and will continue in order to further improve the environmental performance of the plant. With this, I would now like to hand over to Filip to cover the financials.

Thanks, Marc, and good morning, everyone. This first slide recaps some of the key numbers for 2019. Revenues were up 3% compared to the record year 2018. And this, despite the recession in our largest end market, the automotive sector. Revenues grew 7% in Catalysis and 9% in Recycling, but were, to a large extent, offset by the headwinds faced in Energy & Surface Technologies. Recurring EBIT came in close to last year's record number. Excluding the impact of high depreciation charges, recurring EBITDA grew 5%, which includes a EUR 17 million increase due to the adoption of the new IFRS 16 lease standard. The recurring EBITDA margin for the group was stable, with higher margins in Catalysis and in Recycling, in particular. Our recurring net profit was down 5% due to higher financial charges and specifically, interest payments. While still well above our cost of capital, return on capital employed came down to 12.6%, driven almost entirely by Energy & Surface Technologies. From a group perspective, the reason for this decline was not lower recurring EBIT, but the substantial increase in the average capital employed year-on-year following the recent growth investments. The Kokkola assets, acquired at the beginning of December 2019, are also included in the end-of-year capital employed for some EUR 200 million without yet having contributed to earnings in 2019. This next slide puts our group operating earnings and margins in a historic perspective. And we would like to highlight the consistent path of earnings delivery in recent years, reaching a new high in recurring EBITDA in 2019 and consolidating the margin up-trends of recent years, again, despite the challenging market context. This is consistent with our ambition to target profitable growth. When calculating the compounded annual average growth rate since 2015, we reached 11% for recurring EBIT and 10% for recurring EBITDA. Slide 19 illustrates that our 2019 performance was driven by a strong second half as group revenues and earnings recovered from a softer first half. Sequential and year-on-year second half growth rates for group revenues and recurring earnings are strong despite Energy & Surface Technologies feeling the force on its second half earnings of the upfront cost headwinds related to the greenfield site investments in China and in Poland. The graphs on this next slide show that we significantly improved our free operating cash flow compared to 2018, as previously guided. Our cash flow from operations reached EUR 549 million compared with EUR 92 million in 2018. On working capital, we ended the year with an increase of EUR 78 million, entirely situated in the second half of the year. This number reflects a more substantial increase in the working capital needs of Catalysis, driven by the inflation in PGM prices. This increase was partly offset by a decrease in working capital in Recycling that included the release of some inventories built up in Hoboken following the recent fire incidents. Now in view of the sharp PGM price spikes seen since the start of 2020, the net working capital in Catalysis is expected to further increase significantly if, obviously, current prices prevail. I can assure you that managing our working capital remains a top priority for our teams in 2020. Obviously, metal price fluctuations will remain always the dominant driver. CapEx over the period increased to EUR 553 million and is still concentrated on our strategic expansion projects. Some 60% of total CapEx was spent in Energy & Surface Technologies. And the 2 greenfield projects, obviously, took up most of that amount. CapEx in 2019 also included the investments carried out during the extended maintenance shutdown in Hoboken and the expansion investment in Korea for fuel cell catalysts. For 2020, we would, at this stage, guide towards a CapEx level of the same order as in 2019 or maybe slightly higher, but we will obviously continue to adjust and align our plans as much as possible to the market reality. The recent trends of gradually increasing capitalized development expenses also continued into 2019 and accounted for EUR 35 million over the period. Again, most of these assets are related to R&D projects in Energy & Surface Technologies. Accounting for these investments, the free operating cash flow over the period amounted to a net cash out of EUR 39 million compared to a negative cash flow of EUR 406 million in 2018, as you can see, plotted by the blue line on the lower chart. This next slide walks us through all cash flow items, starting from the operating cash flow we just discussed. The combined cash out related to taxes paid and net interests amount to EUR 127 million over the period, which is less than last year as higher interest charges were more than offset by lower cash taxes. The increased dividend in 2018 amounted to a cash out of EUR 186 million. Finally, a key use of cash in 2019 was the acquisition of the Kokkola operations for EUR 188 million on a cash-free basis. At the back of the bridge chart, you can see the accounting effect of the adoption of the new IFRS 16 lease standards on a net financial debt, which is a modest EUR 46 million as we only use limited operating leases. As shown in the next slide, the increase in our net debt of EUR 582 million over the year brings us to slightly more than EUR 1.4 billion of net debt at the end of 2019, which corresponds to 1.9x recurring EBITDA. This includes the new EUR 390 million long-term U.S. private placement debt that was drawn in September. Maintaining sufficient funding headroom to execute our growth strategy and remunerate our shareholders is obviously a key priority. And Slide 23, actually recaps this flexibility with some numbers related to our current, medium- and long-term committed facilities and their comfortable maturity profile. These facilities are complemented by substantial additional sources of funding, including commercial paper programs and bank loans. Finally, a word on nonrecurring items, which had an impact of EUR 30 million on EBIT and EUR 24 million on net profit. They are almost entirely due to a few restructuring initiatives. In particular, the closure of one U.S. site of the business unit, Cobalt & Specialty Materials, as we continue to optimize our footprint where necessary to maintain value creation. This concludes my section, and I'll hand back over to you, Marc.

Thank you, Filip. Before opening the line to your questions, I would like to recap the key messages of this morning's call. I'm proud of our performance in 2019, which was close to the record levels of 2018 against the backdrop of a declining automotive market and a slowdown in EV demand. While we are adjusting with agility to short-term fluctuations in demand, I'm confident that our long-term strategy to be a leader in clean mobility materials and recycling will result in further growth for Umicore. We have taken several major steps in 2019 to strengthen our position and prepare us well to capture significant future growth, and we will continue to execute the strategy with determination. We will also continue to address the issues that challenge our industry, including our unwavering stance on ethical supply of raw materials. Finally, I'm pleased to confirm that we expect to grow revenues and earnings in 2020 despite the adverse market trends that developed in the course of 2019 and continue to prevail today. With this, I would now like to open the floor to your questions. As usual, I would like to give everybody a chance to raise a question. And if you have a follow-up questions -- follow-up question, please place your name in the queue again.

[Operator Instructions] Your first question comes from the line of William (sic) [ Wim ] Hoste from KBC Securities.

Yes. Question is on the wording of the guidance on the catal to volumes. That was changed from tonnage to gigawatts hour perspective. I had in my mind that there was roughly 2:1 rule of thumb to be used. So to be very precise, the guidance you give now, the 60 gigawatts hour by mid-2021 and 100 by mid-2023, does that fully align with the earlier guidance, but then going towards an 18-month delay versus, yes, the initial plans? Is that how we should read this guidance?

Wim, yes, indeed, you are correct. The guidance is fully in line with the previous guidance, which was expressed in terms of tonnages. We decided to move away from tonnages and align with the industry practice of speaking of gigawatt hours, which is what the car producers and the battery cell producers are using as a reference too when they talk about their own capacity or requirements. And yes, it's fully in line. The target that -- and projections that were expressed some time ago in metric tons corresponds to -- considering the mix that we use at the time to do these estimates corresponds to what we're expressing now as 60 and 100 gigawatt hours, respectively.

And what gives you then the -- what is the reason then for now mentioning that you will be moving more towards the 18-month delay instead of the 12 months? Is it market development in China? Is it other reasons? Can you elaborate on that?

Well, we said 12 to 18 months because that was the range, which meant that every point in the range was a distinct possibility. As I mentioned a while ago, we see no sign of EV recovery -- EV demand recovery in China for now and do not expect that to happen before 2021, bearing in mind that the subsidy scheme was supposed to be phased out in 2021 in China anyway.

Your next question comes from the line of Charlie Webb from Morgan Stanley.

Perhaps I could just ask a question around the margin progression in E&ST, just looking at the sequential decline, H2 on H1. Perhaps you can help us understand the various moving parts. Obviously, I see higher D&A, but it's not enough to kind of offset some of the other declines clearly ongoing. So if you could just help us understand the margin progression, H2 on H1? And then perhaps what we should think about as we think about 2020 for the margins at E&ST.

Charlie, so it's difficult to extrapolate from short-term fluctuations and from short-term situations. Clearly, the margins reflect the fact that we are incurring much higher costs, given the higher D&A charges, that's one; the upfront costs related to the construction of our greenfield site; the increased R&D costs; the growing start-up costs for the new capacity; and qualification costs of the new lines. That's one key aspect. One key aspect -- this is the key aspect in explaining the change in the margin profile. The way I look at it is because it's difficult to -- I would say, to compare a quarterly or half yearly margin evolution, the way I look at it is, overall, we're showing -- despite the difficult market context, which I explained a while ago already, in particular, in China with the overcapacity that exists in that leading EV region, I look at our margins, EBIT margins of some 14% and EBITDA margin of some 20% in the depressed market context as being market-leading margins. And there, as far as we have seen, this is best-in-class performance in terms of margin. And this is the perspective that I would offer for you to look at the margin evolution.

Okay. Okay. So sorry, just trying to understand. So as we think about next year, given kind of ongoing investment and everything else, it would be -- would it be right to expect that some of these costs continue? Or would you say -- will D&A step up again, R&D will continue to be a cost, start-up costs will be ongoing? That this is kind of the margin for now as we think into next year, is that fair? Or would you expect some of these things to roll off? I mean, clearly, D&A will go up again, I guess, but in terms of startup costs, perhaps?

I'm not going to be very specific on the margin expectation because it's too early to provide any quantitative guidance for any of the businesses. This being said, as we wrote in the press release that was issued this morning, we expect costs to increase, indeed, because of the higher D&A charges resulting from recent investments and continuing investments in our greenfield production sites, continuing increases in start-up costs and R&D costs. And at the same time, we expect no significant change or recovery or improvement in the market conditions. So these are the 2 dimensions that I can offer to guide you in your margin estimates.

Your next question comes from the line of Ranulf Orr from Redburn.

Marc, I was just wondering if you could clarify your quote on the outlook statement in Page 1. I'm referring to "adjusting investments to take account of evolving market needs." Is that just reference to the update back in April? Or is there other ongoing adjustments being made? And then really, I wanted to ask about E&ST growth in 2021 -- sorry, in 2020. It seems like there will be very limited new capacity coming online. So should we expect growth in line with that? Or do you have the ability to drive asset utilization higher?

And the outlook statement regarding the evolving needs, indeed referred to what we said back in April of last year and it was an illustration of the kind of adjustment that may be required if market needs are changing one way or another indeed. So there is no other allusion that is being made with this statement. And yes, the -- sorry, for the second part of your question, the growth -- the volume growth that we referred to for E&ST is mostly going to come from capacity additions. I do not expect, considering also the constant evolution in the product mix, that efficiencies will drive major changes in the volumes in the short run. So very capacity driven.

Your next question comes from the line of Mutlu Gundogan from ABN AMRO.

Marc, you talk about EBITDA in E&ST being down half year on half year despite EV cash flows being up. Was that your expectation already at the H1 results back in July?

Yes. Maybe I'll take it, indeed, Mutlu. So you're referring to a lower EBITDA margin in the second half. Is that your question?

Not so much the margin but just the absolute EBITDA, was -- I mean, the fact that it was down, was that expected?

Yes. I mean, I think we already highlighted in the July call that we would be facing a cost increase indeed. So apart from D&A, obviously, because we're talking about EBITDA here. But indeed, the start-up costs from the greenfield site is now starting to really come through and has been coming through the second half because we now have, obviously, also the European operations and the preparations for the site coming through. So yes, that was, I think, in line with what we had mentioned in July. The R&D costs, et cetera, we also had visibility on that and that will indeed also continue.

Yes. I mean, there's no business that fell short of your expectations, that's actually what I'm trying to get at.

No, given the market context, no, absolutely. And I would also say because if you look at -- in the question on margins, obviously, we're comparing here with 2018 and the market context is totally different. If you compare the margins with the margins we had in E&ST in previous years, you will see that they are pretty good, especially given the market circumstances we had to face in 2019.

All right. I mean, just a follow-up question on this. Because if you look at the wording on the E&ST in the H1 press release and also now, it's similar in the sense that, yes, we expect volume growth, but there will be additional costs. And then eventually, what you reported was a decline sequentially in EBITDA. So just will this mean that we're also going to see a decline in EBITDA in 2020? Or -- I mean, you don't see a lot about E&ST in terms of absolute earnings in 2020. Can you help us a little bit there?

I think it's a bit too early, Mutlu, to go beyond what we've said on 2020.

Okay, because you are rather explicit on the 2 other segments, so, but not on E&ST.

No. I think you're asking a specific question on EBITDA in 2020. And that, I think, goes beyond where we want to go. And I think it goes beyond the outlook statement we've made, yes.

Your next question comes from the line of Peter Olofsen from Kepler Cheuvreux.

I wanted to ask on your efforts related to batteries, but outside of cash outs. So first, on battery recycling, I know volumes are still a few years out considering the time it will take for engineering, construction, et cetera. What would be the time line in terms of decision-making around the potential industrial-scale battery recycling plan? Could it be that potentially later this year, we get more news on this? And could you also provide an update on your efforts around anodes. Are you already generating some revenues there? What's the progress there?

So no, I do not expect that we will reach a decision point this year regarding the scaling up of our battery recycling activities. I still expect that we will need to be on stream with industrial-scale facilities in the second part of the decade. And indeed, there is quite a bit of a time lag that you have incorporated, I guess -- that you have factored in your question, I guess. As I've typically said that we need 2 years to engineer, 2 years to build, 2 years to ramp up, so we still have a bit of time ahead of us. And we're working on the subject to figure out how to best -- how and where to best scale up. So -- but likely, most likely not reaching a decision point this year. And then in terms of the anodes, yes, we do have revenues from commercial sales for certain applications. These are, however, pretty small at this stage and not of a nature, not of a size that can move the needle in the E&ST segment as of yet. And the development efforts do continue, of course, to broaden deeper technology view and the potential application, indeed.

But this is an area where we may see more significant CapEx in the coming years? Or is that unlikely to be the case?

Well, if we are, I would say, extremely successful from a technical and commercial point of view, yes, that would be the logical outcome. However, given the amount of development work that has yet to be done from a technical and commercial point of view, I think it's too early to be more specific on this one -- to be, I would say, 100% affirmative on this one.

Your next question comes from the line of Mubasher Chaudhry from Citi.

Just one quick one and a follow-up. Would you say that the drop in margins is entirely cost related within E&ST? Or is there a price portion to it as well? And then on a longer-term basis, how would you prioritize between kind of market share and margin for E&ST going forward?

The main effects that we have in the margins are cost related, as we mentioned earlier or indicated earlier. Yes, indeed. I'm sorry, the second part...

And then the second part was just because you talk about retaining market share and I'm just trying to think about how that evolves going forward. Given the choice between market share and margins, how would you prioritize between the 2?

Yes. No, thank you for repeating the question. I'm sorry I missed it in the first place. The -- our priority is and has always been to optimize returns. And we are not obsessed. We're not driven by market share because it's -- in a way, it's too easy to win market share if you sacrifice prices and margins, and this cannot be an objective. Objective is -- our objective is to make this growth business a sustainable growth business with acceptable returns, so that we can continue to generate the means of -- that are required to further invest. So whether we're talking about E&ST Catalysis or Recycling, the common -- one of the common denominators is that we will always prioritize margins and returns over market shares.

And your next question comes from the line of Nathalie Debruyne from Degroof Petercam.

I'm sorry, I'm going to come back on E&ST again because I just want to make sure I understand correctly the building blocks because you flagged the higher cost, that's pretty clear, the G&A, we factored that in. But then you mentioned in the press release, substantial impact from the lower cobalt prices and sourcing of unethical cobalt. Okay, you mentioned that was a EUR 25 million impact, about that, over the full year. But I'm wondering does that include the impact on your LCO business? And if so, could you maybe help us quantify that? And also for the ESS business that [indiscernible] until last year. So that's for the first part and yes, I'm going to have a second question.

Nathalie, let me maybe start with this one if you allow me. Yes, this impact of EUR 25 million -- estimated impact of EUR 25 million includes the impact on LCO. LCO is one of these high cobalt-containing products, which is facing very significant competition from products containing unethical cobalt. So this is one of the issues that I flagged last year, indeed. And on top of that, it's also worth to remind everybody that LCO sales were also down because of the excess inventories across the value chain. And we saw a general movement in the industry to reduce these excess inventories, in particular, towards the end of 2019.

Okay. And for the ESS business, is there any chance that you can give us an order of magnitude of the impact it had on the 2019 numbers?

No, we're not going to detail that too much. What I can confirm, which compared to what I said last year, is that the impact is material enough to be mentioned.

All right. And then maybe, if I may, and then I'm going to stop. I just wanted to have a bit of an update of your hedging strategy, especially for Recycling. So I just wanted to have a bit of an idea of what -- of the portion of metals exposure of 2020 that is already hedged?

Yes, so in the press release, what we've written is that we have locked in more than half of our both 2020 and '21 exposure for gold and palladium. So that's a significant portion. And then also a significant part of our 2020 exposure for platinum. Now it's significant versus more than half, it means like significant sort of platinum portion is somewhat less hedged than palladium and gold. But -- so it means that, indeed, for -- especially for 2020 and also 2021, for gold and palladium, we have already a good portion of hedges in place, more than half of our exposure, which obviously means that yes, it caps in a way part of the upside that you see in the recent price spikes, specifically for palladium, obviously, because these hedges have been entered into, I would say, during the course of 2019. And you know the strategy is really to create visibility at an attractive level. What I would, just to finish on the hedges, say is that on rhodium, because that's also a metal that has been increasing a lot in 2019 and especially recently, there, we have no hedges so -- because rhodium cannot be hedged. It's not a paper market. So for palladium, more than half hedged as well for gold; for rhodium, no hedges; and then we have a few additional hedges for other metals, but those are less significant.

Your next question comes from the line of Alex Stewart from Barclays.

I have a technical question. If you build a new asset, and it takes you 2 years to build the asset and then ramp it up in year 3, at what point does the cash you've invested, so the capital on your balance sheet, get allocated to the divisions? And at what point do you start depreciating the asset? Do you depreciate it immediately when it ramps up? If you could give us some sense of how the accounting works, it would be extremely helpful.

Yes. So the depreciation starts when the commissioning. So when we use the assets, that's the simplest way to put it. So once we are really in commissioning and commissioning when it's really, I would say, operational commissioning, then we start to depreciate the assets. So what you see in terms of D&A and increase in D&A in '19 is really still largely related to the investments we've done in recent years and partly the Chinese plant for part of the year. And then -- so the European plant, that increase, you will actually see mostly as of 2021. But obviously, the Chinese plant will continue to be depreciated in 2020.

And just on the invested capital portion, at what point do you make the asset live rather than work in progress?