Okay. Good afternoon. Welcome to our 2019 investor update presentation. Thank you for joining us here today in person or via the webcast. Today, we have Ivan Glasenberg, our CEO, Steve Kalmin, our CFO, and Peter Freyberg, our Head of Industrial Assets, giving an update on our prospects for the next three years. I'll hand it over to Ivan.
Okay, good afternoon. I'll first present, then Peter, then Steve. To give you an idea, our investment case, we always talk how we differentiate Glencore for the other mining companies and the advantage we have, et cetera, and where we see the markets and what we see going forward. We'll talk about all these issues, then we'll go in more detail. Peter will give more detail on the production and Steve on the financials. If you look at the markets today, where do we see the markets today? What's happening in the markets? It's clear the markets are pretty well-balanced. The markets look very tight, and we're seeing a large amount of destocking of the inventories, whether you're talking LME, SHFE, off-LME, et cetera. We've seen inventories run down in most of the commodities which we handle.
What we do see happening is that getting the commodities out of the ground is in more difficult regions and the easier stuff is disappearing. We've got to go to Africa, we've got to go to Russia, we've got to go to more difficult regions to get the commodities which we produce. We think supply is starting to tighten up because of the difficulties in mining in those regions. We believe we're well-positioned for key future growth because of the type of commodities we have. Urbanization, another 2 billion people coming into the world over the next 10 years, et cetera. Electrification, the mobility, electric vehicles, and the decarbonization of energy, which is occurring in the world today.
We believe with the demand that we see with these different scenarios and the supply side, which I spoke about, we believe markets should start potential, which is a strong potential for our commodities, and we should be moving into more positive territory. Our business, we've got a unique combination of our assets and the marketing, which we always talk about. We've got a large marketing business. We talk about it, the amount we make in the marketing business, and we've always said with an EBIT of $2.2 billion-$3.2 billion, and it continues to be strong, and it assists us in ensuring that we're marketing our products which come from our mine in the best available manner. We got large, long life assets. If you look at all our assets, they're first quartile assets. They're low cost producers, and they have a long reserve life.
If you just take example, which people seem to ignore about the quality of our assets, and Peter will talk about it later, if you just look at copper today, we got three of the best copper mines in the world today. If you have a look at Katanga, and we'll talk about Katanga in detail, with its cobalt and the benefit of its cobalt credit, one of the lowest cost producers in the world when it ramps up to 300,000 tons. If you look at Collahuasi operating well. It's a mine that will produce 500,000 tons. It has the potential over time to build up to 1 million tons. It has the reserve base to do that. We all know Antamina is one of the lowest cost mines.
If you look whether you call it a zinc mine or copper mine because of the byproducts that it gets from the material. We have a great set of copper assets. If you look at our zinc assets across the board, our coal assets, first quartile, long-term, large life reserve mines. We've always said we've also got growth, which are internal growth projects. Peter will take you details through some of those, where we can expand the production of our assets across the board, and we've given a detailed chart on that. The marketing, as I said earlier, has been counter-cyclical. We know when commodity prices have fallen in the past, the marketing has still been strong and the marketing has been within the $2.2 billion-$3.2 billion range. When markets are tight and commodity price is higher, we perform better.
Steve will talk you through some of the marketing prospects going forward. The balance sheet. We've said we're going to run a robust balance sheet. We're going to aim to have a net debt to EBITDA at 1:1 ratio. We got a maximum limit that we will have on the debt of $16 billion, which Steve will give details about later. We'll continue to maintain a strong, robust balance sheet going into the future. It's highly cash generator business with CapEx coming down and going down to sustaining CapEx in the future, down to $3.7 billion, roundabout there. You have a look at the free cash flow we'll generate. At today's spot commodity prices, we'll be generating roundabout $4.4 billion free cash. That's pretty significant going forward. What is our big creating value in our business?
We have got an experienced management team. As you know, most of our management team have been in the company for a large length of time. Most of the people within the trading business have been 15, 20 years there. We've also got good asset managers around the world who are taking care of the assets. Where we also like to say that we differentiate a bit in the market, we focus on value creation, therefore, we're not all about growing and building mines just for the sake of building mines. If we believe mines don't give the right NPV, there's no reason to reinvest the capital back into the business. We'd rather return cash back to shareholders, whether we do share buybacks or dividends. The idea is not just to growth for growth's sake.
We're very cautious where we put our money back into the business, where there's acquisitions, whether it's expansions on brownfield expansions, greenfields. We're always concerned about greenfields. We're very focused on getting the right value for our cash and rather returning it to shareholders than putting it back in the ground. Flexible business. We adapt to change quickly, and you've seen we're very much focused on volume. We're not there about just creating value. We're all about value, sorry, not volume. You don't see us expanding mines, as I said earlier, for the sake of just having a bigger volume. You will see us pulling back on volume. We get a lot of flak or people say we didn't hit our volume targets, and Peter will talk about that later.
A lot of the reason we don't hit sometimes is not our volume targets because we took a decision to cut back on volume. We'd rather leave the material in the ground than pull it out when it doesn't create value and it's not adding extra value to the asset. You've seen us do it in the past, and it's a continued thing that we do in Glencore. We're always assessing all our mines, and we're saying, "Does it make sense to increase the volume?" The material is going to stay in the ground anyway. No reason to pull it out when it doesn't create value. You've seen us do it. We did it in the past in zinc when the zinc price was $1,300. You remember in 2015, we cut back production, and we know the effect it had on the zinc price.
We believe rather pull it out of the ground later on, which we're doing today, when the zinc price is towards $2,400. The same, you saw us do it in coal. We've done it in coal over the years many times when the market was weak. Rather leave it back in the ground. You're seeing us do it continually in Colombia, whether it is at the Cerrejón operation, whether it's at the Prodeco operation. If we feel it does not make economic sense, if the markets cannot take those tons into the market and it won't create value, we pull back. We've done it at coal, and more recently, you saw what we did in the D.R.C. at Mutanda. We decided to shut the Mutanda mine. It didn't make economic sense with the cobalt price where it was or the copper price where it was.
Put the mine on care and maintenance. It, therefore, will be dug out of the ground. As you know, we'll talk about it. Peter will give more details about it. In two, three years’ time, when it makes economic sense, we'll put Mutanda, when we study the sulfides, we'll then decide when to bring it back into full production. Meanwhile, that's very important within Glencore. It's all about value rather than volume. Rather leave it in the ground, pull it out at the right time, and it may affect our volume targets, but from a value point of view, it makes clear sense. With that, Peter will talk about the volume, the details of the mines, the operations, how they're performing today. I leave it to you, Peter.
Thanks very much, Ivan. I'll take us through the operational side of the business. The outlook over the three-year period is essentially flat production in terms of copper equivalent tons. What we are seeing is that we have got changes, obviously, in our zinc and oil and increases over the period that I'll take you through. This offsets some of the reductions we're seeing, such as the shutting down of Mutanda that is currently underway. Just in terms of the growth that we talk about within the presentation, obviously, we're seeing the benefits of some of the investments that we've been doing over time with Katanga ramping up. Finally, that will be coming back on stream early next year. I'll talk a bit about those details, certainly with where we are going with Katanga at the moment.
We're seeing some reasonably robust growth in zinc, driven by a couple of areas, but one of them, again, being something that we've invested in to offset future declines that we have at some of the operations within Kazakhstan and within South America and Canada. Coal, we're seeing fairly flat, although we do have a project coming in. It has been held up a bit with permitting, but coming in nevertheless. Nickel, we will talk a bit about what's happening at Koniambo. I know that's one of the focus assets that people are interested in. The declines are pretty much where we expect and what you've heard about before with Mutanda coming out of the system, and obviously some of the late-life assets such as Matagami and Iscaycruz in Peru. What we are seeing is the benefit of some of the long-term work and investments that we've been doing.
We are seeing how we respond to the cycle. Again, Mutanda is a good example of that, having determined this year that in the current market, and partially as a result of the new fiscal regime and tax structure within the DRC, that that business wasn't giving us return. It makes sense to park that up, notwithstanding the fact that it is a source of a large volume of cobalt. Again, value ahead of volume. Really what this does demonstrate, and when Steve takes you through the earnings, is the power of this diversified model that we have and its ability to give a sensible return through the cycle if we don't just chase, even in the bottom parts of the cycle, if we don't just chase the volumes when they lose money.
I'm not going to go through the details on this because we'll be talking to it on each commodity group. Obviously, the intention areas will be copper. Zinc is quite interesting to see how that grows over the next several years. Also our oil business, which is showing some output growth there and value growth, and Steve will touch on the earnings benefit of that. Talking about the focus assets. These are the ones that we spent a fair amount of time discussing at the last presentation. Just a quick update on Katanga. We have essentially achieved everything we said in the last meeting, and we continue to remain on track to deliver. We're pretty comfortable with the 235 target that we set ourselves earlier this year on the copper side, having understood some of the challenges that we had.
We're hitting that on a very consistent basis now. The cobalt, which previously had given us some challenges late last year, early this year, is all on spec. We did have some issues starting up the dryers. The first dryer is now in operation, running at about 60%. The second dryer will kick in early next year. We don't see any issues with achieving our targets that we set ourselves for next year and that have been well-publicized in terms of producing sellable on-spec cobalt. The acid plant, we signaled last time that it would be late Q1 for a startup. That remains on track for that. The electrowinning plant, particularly electrowinning plant two, where we've got a major anode replacement program, on track as we said previously. Katanga is going real well. We've got a new management team in there to some extent.
We've brought some extra people in. We're now very comfortable with the structure that we have in place and the skills, and pretty good about where that project is going. Mopani, as you're all aware, we had to enter into an early rebuild on the smelter. That is tracking well. The rebuild should be finished towards the end of this year, and we expect ramp up in the first quarter next year. Koniambo, as I think everybody is aware, we had a pretty tough first half. Second half has been very consistent operation. We have seen the best power plant performance that we've ever had at Koniambo, with almost all of the power generated through the coal units, and they are performing very consistently now. In fact, we've had 43% improvement in power generated into the met plant.
Smelter, the improvements there, we've had, again, significant improvements on furnace online time in the second half, which has increased by over 50%. Obviously, the ferro-nickel production goes with that. Noting that in the first half we had some scheduled downtimes. We are getting consistent operating days to the extent that we've had our best continuous 60-day period in that plant in the history of Koniambo. Together with that, you obviously get other improvements such as improved recoveries, where we've seen a 14% improvement in recovery in the second half compared to the first half.
Although our nickel output isn't where we want it to be in the long term, at this point in time, we are trending at levels that allow us to achieve the sort of targets that we've been setting ourselves for next year and that appear in the guidance that you will see later on. That allows us to continue to build in terms of reliability and steady operation, which means that over the next few years, we're going to be targeting around 30,000-40,000 tons per annum. We continue to target 50,000 ton annual output as a long-term target. Really, as far as I'm concerned, sound progress with the focus assets. Certainly my team fully understand the importance and the value associated with delivering against these assets, making sure these assets perform. Looking at copper guidance, again, reasonably flat through the period.
We have separated out the African assets from the rest of the world. There is a modest decline, which going from this year to next year is fundamentally Mutanda. We see the result, I guess, of a steady state operation coming in those African assets, particularly Katanga. As it ramps up, we expect it to reach its steady state sort of levels towards the end of 2020. As I said previously, that's on schedule. There are some copper tons that do come out of the system, but they're principally at non-copper managed assets. You've got a reduction in copper operations such as the Integrated Nickel Operations in Canada. You've got a small reduction of copper coming out of Kazzinc and obviously mines like Kidd Mine as well.
The cobalt guidance at 43,000 tons this year reflects Mutanda is still operating and then we are, as you probably have seen in the media, we're in the process of actually shutting Mutanda right now. You see the decline in cobalt tons coming out. The 29 is mostly achieved at Katanga, the 29 guidance plus or minus four. There's obviously, as you'd be aware, some tonnage coming out of other operations such as Murrin and a small amount out of the Canadian operations as well. Looking at coal guidance, we've revised this year down a little bit. It's principally come about because of a reduction within some of the Colombian operations, and we've also had to have a safety outage at one of our South African mines. We do see the tons decline into next year.
That's fundamentally as a result of some higher ratios at some of the Australian operations and a reduction at one of the underground mines in Australia where we have two longwall moves rather than the usual one longwall move. After that, the profile flattens out. Within that profile, there is also the United operation, which is a 50% joint venture, of which our share is ultimately 2.5 million tons over the period. It really starts producing towards the end of 2020. Each area, generally steady sort of production. As we'll talk about in the marketing later, and there's some slides on it, in a world where it is difficult to permit mines and where we are seeing coal demand growth, you'd expect that the returns on that tonnage should start to look quite interesting.
As I foreshadowed earlier on, zinc has got quite an interesting profile. You'll see there's quite a bit of variability across our diverse portfolio that we have there. We are seeing. Included in that is the Antamina zinc, where we're seeing some very significant increases over the next couple of years. It's just the nature of the skarn deposit. You go through these periods of high zinc. In fact, there's a little bit more zinc than copper in the next couple of years. That then declines again. That's the top of that bar chart. Also, we are commissioning our Zhairem project in Kazakhstan. That ramps up next year. Then grows very rapidly.
It produces a very significant amount of zinc in 2021, then trends towards its steady state sort of output of around 150,000 tons of zinc per annum from 2022, 2023 onwards. Australia, very steady output at McArthur River and George Fisher, Lady Loretta complexes. South America, something of an up and down. We have not included in those numbers, you'll see it in the footnote, our Volcan business, which is quite an important tonnage for us. It isn't included in that. Within our South American operations, we are seeing a bit of a decline towards the end of the period as a couple of the mines, such as Escobal, Iscaycruz, come out of the system. We have Volcan, which is a very important resource for us that will fill those gaps over time. A really interesting zinc story.
Nickel guidance, I talked to what is happening at Koniambo, as I said, I'm actually pretty pleased with the steady state that it's at. We're not happy with its absolute level of output at this point in time, we will be in the 30,000 to 40,000, the lower end of the 30,000 to 40,000 range next year. We will see that grow over the period with the ultimate target, as I said, being around 50,000 tons. We are getting great performance out of the power station, the met plant itself is starting to perform in a predictable and perhaps more reliable way. INO, you'll see that integrated nickel operations, we actually have tons coming down. We've known this for some period of time, it's a result of the Sudbury operations volumes coming down, which is something that we've seen for a while.
However, because of decision-making that was made some years ago as to when to ramp up the Onaping Depth, we have a bit of a decline, but then it picks up again. In fact, in 2020, just beyond 2022, we see the nickel output in the integrated nickel operations normalize again. The projects that we have there with the two new Raglan operations are on schedule. I sat with that team about 10 days ago and went through the projects in Canada, and we're satisfied with the way that they are heading. Murrin Murrin continues to be a good performer. Its output is very consistent through the period, and it continues to have that sweetener of a little bit of cobalt in there, which when we get a good cobalt price, is very beneficial for that operation. Oil.
This is, Steve will talk later on about the earnings associated with this E&P profile. We are building up from something that was in terms of EBITDA, $150 to over $650 over this period, which is pretty exciting for the oil business. It's associated with some decent work that's happened in Chad in terms of the Badila oil field that is sitting at the bottom of that bar chart. From the work that we've done, we're seeing some steady growth in volumes there. What we are seeing in Equatorial Guinea is actually the oil equivalent of the gas that we're producing. We are in the process of bringing in the gas project in the Alen field, we expect that will be really kicking in and starting to produce from quarter one 2021. A very, very positive project with very short paybacks.
It's something that's been a good development for us. Obviously Cameroon is a little bit of cream on top of that with the volumes increasing in Cameroon as we get the Bolongo field producing. Excuse me. Steady over the period, in terms of volumes. We do need to look into the future in terms of what options we have to increase output when the markets are ripe to do that. As you know, we don't like to bring the volumes on until we think the market can accommodate those volumes without damaging the rest of our business. We do have a tremendous suite of opportunities and options across our mining portfolio. You'll see the projects listed there. I think you'll all be familiar with the sorts of resources and reserves that we have in those businesses.
Coroccohuayco project, for example, is very well-positioned so that we can maintain levels of output at our Antamina operation in Peru. That typically runs at around 200,000 tons of copper, and Coroccohuayco is there on its boundary. As the grades go down and Antamina provides us the option to obviously invest there and maintain those sorts of levels of output. Ivan spoke earlier on about the Collahuasi project, or opportunities. That is a reserve and a resource that just goes on and on and on for decade after decade. Obviously as the market demand increases, and we're talking a bit later on about electric vehicles and decarbonization and the impact that has on demand for copper.
Having projects such as Collahuasi sitting, or it's not a project, it's a highly productive, highly profitable mine, but having that optionality sitting within Collahuasi to grow that business, is of massive value to us as well. Also obviously want to talk about Mutanda. We've taken that offline. We are progressing the studies there. They're at the pre-feasibility stage, we will decide how we, and what we do with that, when the time is right. There are a number of factors that we will need to consider, including cobalt supply demand, but also the tax regime and progress that's been made within the DRC as to whether or not that facilitates bringing that asset back sooner rather than later. Nickel, we are progressing those projects. The Nickel Rim Deep, which is technologically a very advanced operation. It'll be the first all-electric ore mine underground.
It is reaching depths and temperatures that other mines don't. We remain very confident that that is a great project and allows us to take our INO complex back to the levels that they previously at in terms of nickel output. Obviously the two Raglan projects. There's more upside around those Raglan projects in terms of the ore body there. It's a tremendous ore body and is a very important contributor to that region. zinc, I touched on Volcan earlier on. It is an incredibly prospective area. We have a wide range of projects within Volcan that are being studied. There are a couple that are already being brought on in terms of small open pits, over the next year or so. That remains an interesting prospect for us in terms of growing our interest in that business, if that's a doable venture.
In Kazakhstan, we are currently getting to the last six to nine months of the construction around the Zhairem project. That comes in next year with not so big tonnages next year, but the year after that starts bringing some very significant tonnages. We also further have a number of options there where we can grow our zinc business as demand comes about. In coal, we've always talked about the fact that we do see demand continuing at current levels, particularly with growth in thermal power stations in Asia. We have a fair amount of optionality there to bring projects in to meet that demand as it arises. Some of them are continuations of existing operations, such as Glendell and Mangoola and Bulga, as well as the HVO extension. We've all seen what the power of those reserves at HVO are.
Also more recent prospects such as the Valeria project that we acquired from Rio Tinto about 18 months ago. Probably the most important thing I can end on, in terms of the operational update, is where we are with safety. The last time I spoke to you, we said that we were in a bad place and the results were unacceptable. The reality is, we're still in that bad place, in terms of the number of fatalities that we've had year to date. 16 in all. If we look at the details in there, we can see that the majority of them are in our copper African operations. We've also had a significant number of fatalities in our zinc business. Our coal business has had a fatality, as well as our alloys business, both of those occurring in South Africa.
This is something that has shocked the business to the core. We know that this is something that we have to deal with, and we are dealing with it. In terms of the African operations, right from when we started having a problem in Zambia, we brought in a very significant team to work through that business. Certainly, if you look at Mopani today, it's a very, very different business to what it was when we had those fatalities. It continues to have challenges. We have looked at the actual management structure within Mopani and made very significant changes there, in terms of how accountabilities are structured, in terms of how training is done. We've changed the rosters on the mine. We have done a major review of the contract operations within that business.
We do believe that we have set Mopani on a course for it to be successful in terms of its safety outcomes. Similarly, the DRC, we've got a lot of new people on the ground there. We've brought in some additional resources to help us work through the problems that we have there to make sure that we fully understand every aspect of those operations, all of the hazards associated with it, and that we eliminate those issues. Within our African operations, we've had very significant interventions. I personally have been within those operations six times this year, and we will continue to maintain a senior management focus on those until they are absolutely right. As far as zinc is concerned, currently we have a team of 30 people working through our Kazzinc operations, addressing some of the, what we think are shortcomings there.
It's proving to be a very cooperative and very structured and hopefully very successful process. I was there about three weeks ago. I'm there again next week working with the intervention team plus the Kazzinc management team. We are going to eliminate these fatalities. We know what we have to do. Where we have safe work working well in our business across our suite of 150 assets, it works very, very well. We have a very large number of assets that run year after year, large suites of assets that run year after year without serious incident. We do know how to do it. We just need to make sure that these focus assets, the challenge assets, that we've got it in place and working. We're not in a good place when it comes to safety. We do know what we have to do.
We are doing it, we will turn it around. Steve?
Hi, everyone in the room and those that may be listening or watching through the web. Peter's taken a few slides that I historically would have covered in this presentation. It's lighten my load, we'll just jump straight into some of the financial building blocks and the way we see the business, in terms of closing in on 2019, more importantly, as we look towards 2020. As in previous periods, we've tried to give you some expectations on some cost buildups for 2020, and I can give maybe a few points beyond that. Copper was the key one that we clearly had with some underperformance at Katanga Mopani in a cost sense, and in the delayed sales of cobalt. All these things, of course, with applying cobalt as a byproduct towards the copper cost that we have within that business.
In terms of the graphs, you can see at the bottom, 2018 was obviously actuals. The 2019 H1 update, that was the full-year guidance update that we gave back in August. We're not going to be freshening that up now. We don't see any great need to do that at the moment. We'll obviously, in a couple of months, we will see you all again for actual delivery in respect of the 2019 performance. The 2020 is then predicated on the production profile that Peter went through, as well as what our cost expectations are for 2020. We think we will be turning the corner reasonably materially on the copper business as we go from 2019 into 2020. You can see hopefully that's a peak of 156, and 120. Ex-Africa, we've been pretty stable at that 101 down into the 80s.
That's the rest of the business, including Africa. It's peaking at that 156, reducing to 120, with some further downward pressure on that unit cost towards 2021 once Katanga is towards the 300, and we're normalized our level of cobalt sales and the cost structure is again more optimized around particularly the acid plant coming in through the first half next year, which will significantly reduce the cost structure of that business as well. Later on you'll see based on the 1.3 million tons copper at that cost structure, we'll be doing at spot prices today on copper and various byproducts around $3.6 billion of the EBITDA in copper, which will be a pickup on this year, notwithstanding the reduction in Mutanda.
Those were not contributing cash flows from both a free cash flow or an EBITDA, given the copper, cobalt, and cost structure and tax structure within that particular business. Nothing's changed in terms of overall cost structure. It just looks spreadsheet-wise like you've got some tick up in unit cost per ton of zinc produced. It's just we're adding quite a bit of zinc. Mathematically, that zinc is not proportionally bringing the byproducts like the gold we have as well. You obviously have the Vasilkovskoye operation, which is steady at around the 500,000. The growth in zinc tons, as Peter mentioned, primarily coming Zhairem, a little bit of lead, but essentially a zinc operation. The Peru, the restart of Iscaycruz as well. Quite a big pickup, 150,000 ton you can see on the zinc side.
You're just throwing less byproduct, absolute volume across bigger tons. You reduce the structure, but the overall shape of that business at the 2020 would produce around $1.7 billion of EBITDA in the zinc business as well. Nickel at the 125,000 tons flat cost structure. They're stable unit cost, a second quartile cost structure as we see that particular business as well. That producing EBITDA around $0.7 billion at that level as well. Going through a little bit of a dip in INO, exactly as Peter said, through the depletion of some of the older projects and the new investments we're doing in the Onaping Depth around Sudbury, as well as the Raglan extensions should see that. It's just outside the forecast period when you see that additional pickup beyond into 2023. The coal is the dip next year, as Peter said, down to 135.
That's a bit coming out of Australia and a little bit coming out of Colombia. Fairly stable cost structure. The way we see current net pricing, it's quite a complicated model that we throw all the ingredients into of a Newcastle. You've got stable domestic pricing in some countries. You've got the coking coal byproducts. You've got the various quality differentials. You've seen some closing of the gap between some of the European, South African FOB pricing in the last three or four weeks. I think there's on some benchmarks, we've seen some $30 per ton increases, particularly API 4s and the likes coming out of Colombia.
That's all thrown into the big pot, and you have an average realization of around 70 or so at current pricing, giving a $23 a ton margin across our 135 million tons, which is around $3.1 billion of EBITDA into 2020. I mean, where we have managed tonnage around the overall market, where we have reduced tonnage, generally it hasn't come at the expense of cash flow. On the contrary, it may have been converting negative consumption of cash flow into a more positive cash flow environment. You've seen it.
I mean, we don't give the numbers over here, but ferrochrome also in South Africa is a good example of that as well where you can see earlier on in the profile dropping about 100,000 tons or so of ferrochrome, and that is very much a response to the current weaker ferrochrome pricing environment as well. If we move into CapEx as well. A few things have changed, either the footprint, some of the accounting. I've tried to do a bit of a like for like, but overall, fully loaded CapEx around $5 billion over the next three years, 2020 to 2022. That's up from around $4.8 billion when we were here 12 months ago, looking at average from 2019 to 2021.
In the middle of the chart, I've tried to show what's more of a like-for-like comparison against last year and what's either additional accounting and/or footprint within the business as well that would explain some of the movements. We've got an average, effectively, $0.2 billion per annum uplift over December. There are various ups and downs, and I'll talk about some of those. Essentially also the Astron business, that's the downstream refinery and marketing and distribution down in South Africa, which we closed in May this year. That wasn't part of any of the numbers that we had last year. There was an expectation that that would close. It was taking a while, but it wasn't baked into any of the numbers at that particular point in time.
We can also come and show what the cash flow and the EBITDA profile of that business as well looks going forward. The impact of the new leasing standard. You'll all be familiar in your various coverage universe. I think we've been modestly affected by it compared to some other big oil and gas companies and the likes where pretty much all your previous operating leases have now been capitalized. Similar type transactions will be similarly treated going forward as debt-like and CapEx obligations going forward. That itself is around $200 million, $0.2 billion for periods 2019 and 2020 of things that were already baked into our business plans that are now coming on as effectively CapEx and capitalized S now. A good example of this, for example, in the next year or two, we'll have to replace a big icebreaker ship up in Raglan.
Given the territory up there, it needs to work. This is a piece of equipment when you do buy it has a life of about 20 years. This is a typical leasing type of vehicle that you have. We've been running with operating lease expense. That's a piece of kit that's going to have to be replaced in the next 12 to 18 months. It's somewhere in the order of $100 million. That's got a change in geography in terms of where it goes. That's essentially in the sum of that sort of infrastructure type things that may have been lease led in the past. We've also got some which would be no different, I would imagine, within the industry itself.
We've looked long and hard around all the different 150 sites and looked at where we need to allocate and spend a little bit more money around even being more conservative and robust around tailings dam structures and reinforcements, and just building in more and more buffers around some of the technical probabilities and statistics and events that goes into. These are being led through the industry, they're led through Canadian standards. There's a variety of the industry moving towards trying to obviously get the market comfortable around that residual list that obviously looks. Yes, there's some additional money that's been built in and allocated within the budget during that particular period. Those are all negatives in terms of some extra Astron, there's some extra leasing, there's some extra TSF that's gone in there. Less on the other side.
There's been offsetting savings just around project, not using all the contingencies, some FX savings that have come through. Mutanda, for example, obviously in scope has obviously come out of that. Looking at 2019, like for like, we would have been $4.6. There's an extra $4.4 that's come in around capitalization of leases and obviously the Astron. Next year as well will be $5.5 fully loaded, $5.1 on a like-for-like, $4.7 and it does go on to 2022, which is the period currently will be beyond the current investment cycle. That's not to say we're not going to come and initiate and talk about some of those other growth options at the appropriate time that Peter spoke towards as well.
We're also doing a little bit more work around some of the various studies, drilling feasibility-wise, that does allow one to push the button and to monetize some of these projects. There's a bit more that's gone into some of those studies, be it PolyMet, be it El Pachón, Mount Isa Copper. There's a bit more money that's gone into overall drilling and studies across that business. On the far right, oil, hopefully it will generate a return. We think that the returns, the RRs, the paybacks are sufficiently robust, particularly in the E&P side and certainly over time on the Astron side. What does the potential EBITDA of that industrial business look like going forward from a rounding error?
Clearly in 2018 and even this year in 2019, to at sort of $65 a barrel, you get over $650 in 2022, and with a exactly at the time that your CapEx profile also then peaks and starts turning south, you have significant free cash flow generation coming out of that business at that particular point in time. The marketing part of the business, unfortunately, not going to tell you too much or update too much on that front other than suffice to say, we'll be within the range, even after the cobalt mark-to-market adjustment that we spoke about back in the June number. That was a $350 million non-cash amount that was announced. Of course, with the pickup in cobalt prices, particularly on the back of Mutanda going into care and maintenance, it sort of came back to about $18. It's now back also into the 16s.
Some of that $350 would not be that number. It would be a lesser number. Even post-cobalt, the fully absorbed number as we go through to the end of the year will be tracking within the range as we go forward. We were nine, I think $950 net of cobalt for the first half. The combined effect of all that will move us into the range as we speak as well. In the absence of anything that you know we know, the business itself, the performance of the overall business, we think that market conditions for 2020, middle of the long-term range is a fair sort of anchor for now as we move forward in terms of the marketing range as well. We'll come in a month or two, and we'll tell you where we finished up for the year.
In terms of building blocks, cash flow generation at spot. Despite some tonnage declines, be it in chrome, be it sort of pulling a bit of Colombian coal, whether it be some of the other business that hasn't come at the expense of cost structure and margin generation. That's exactly the thesis both Ivan and Peter spoke to the value of a volume thesis. This is building up the way we would see a 2020 illustrative cash flow generation. EBITDA spot price is $12.4, made up of the copper, zinc, nickel, coal. The other is in the $0.4, which is a combination of oil, ferroalloys , other bits and pieces, less the $350 million or so of SG&A. The marketing EBITDA, that's midpoint EBITDA at the sort of $2.7 that we just spoke about. $2.7 cash taxes, interest, and other coming through based on that generation.
Obviously, interest rates have been coming down a bit. We've seen some reduction in funding costs from just below sort of 3.9 to more like 3.7 or something on an overall blended between our fixed and floating exposure that we have. That's CapEx, cash CapEx at the $5.5 less the $200 of capitalized leases. That's going to be the cash generation at spot prices that we go through. In the appendix on page 31, I'm not going to go there on the slide itself, but we have provided some of the sensitivities against those assumptions as we stand at the moment. Top four would be copper, 10% sensitivity against those numbers, around $700 million $0.7 against current prices. AUD interestingly makes it at number two on the podium, around the 10%, followed by coal and zinc are the top four as we go through those.
We will just finish up a Page on balance sheet and capital structure generally as well. As you know, this year, 2019, distributions and buybacks, there was the $2.7 base distribution, $0.20 paid in respect of 2018 cash flows. The $2 billion share buyback program is nearing completion, will be done by the end of the year. Again, Appendix 29, I think it was, shows the latest share count and the projection towards the end of 2019. We would have bought back a little over 7%, I think, of the overall shares since the share buyback started July last year. The all-in purchase price, if you've been tracking that, is less the dividends that have obviously been saved through foregoing that on the shares would roughly be sort of at the money, sort of circa about where we bought it.
2020, you know the minimum distribution policy, that's obviously anchored in around being able to model something. Given the free cash flow generation, $4.4 billion, that I've just obviously run through at the current prices, we'd be seeking and in the absence of any updated and further news around macro and developments that, of course, we would reflect the latest news and understanding and projections of the business and prospects of the business going forward in February. We'd obviously seek to match 2019's distribution of $0.20 when we get to the end of the year, being comfortably well covered, $4.4 billion cash flow today covering that $2.6 billion, 1.7 times. We have our chairman here in the audience. He'll be presiding over a meeting in a couple of months to clearly discuss and opine on what's the appropriate level for 2020.
Net debt maintained at $10 billion-$16 billion range augmented. I have put a footnote there at footnote one, is reflecting the new leasing standard, industrial leasing. I am not looking to make any adjustment to that particular range. The icebreaker I mentioned, obviously, any fleets that we may choose in the future to lease or via contractors to bring onto the balance sheet, that will be certainly covered within the range. I would be looking to exclude within that range, not for the purpose of net debt, EBITDA ratios and whatnot, but the marketing leasing that will come onto our balance sheet that is around $0.6 billion at the end of October. I do not know what it is going to be from month to month because every time we go out there and charter a vessel for 15 months to handle. This is within our trading business.
Maybe $10 million a month, whatever you're paying, suddenly there's a $30 million lease liability or something for very short-term things. You can see there was $0.6 billion. It's primarily chartered vessels and storage facilities within the business, and more than 50% of those commitments expire within a period of two years. These are very short-term, sort of consumable, logistics type aspects of our marketing business that can go down to zero. It can go to $1 billion. These things is clearly not part of your long-term capital structure. Just for the purpose of 10-16, I think, the guidance should be seen in the context of normalizing for that. At the net debt EBITDA level, we're 1.24 at June.
As Ivan said, we'd look in the current period to try and move that more and forwards towards one times as opposed to not letting it go up. It's not die in a ditch sort of one times. If it's 105, 107, 117, these are all pretty conservative numbers as you would all attest to. Non-core asset disposals. Yes, the billion-dollar target is still something which we are chasing internally from non-core long-term asset monetizations. We've completed about $0.3 billion. Not much of a sort of fanfare around that stuff. It's a whole collection of various shipping assets and other infrastructure, smaller stakes in listed companies and likes and some other monetizations, but it does add up. It was around $0.3 billion and a variety of other areas in the same infrastructure, shipping, long-term loans, long-term receivables, VAT.
There's all sorts of things from which could ultimately liberate and unlock some cash that's part of that asset that clearly within a modeling or an asset sense is not something that you'd otherwise be accounting for building into the underlying cash flow of the business as well. With that, we'll let Ivan wrap it up.
Thanks, Steve. How is Glencore positioned? We think we're well positioned. As I said earlier, we got the right commodities. I don't want to repeat on this slide too much. As I said, and it's been proven over the last two years, if you look at both nickel and zinc, there's been a deficit. Because of that deficit, you take nickel over 2017, 2018, there was about 150,000 ton deficit in 2017, about 180,000 ton deficit in 2018. 2019, a bit less, around about 60,000 ton deficit, we believe. If you look at nickel, we've reduced the stocks considerably over the last two, three years, and we're down to 15 days consumption, which is extremely low. Very low for zinc.
As you see the graph, there's been a zinc deficit over the last two, three years, and we've run down to three days consumption sitting in inventories around the world. The same goes for copper. Now, if you look at copper, what is interesting, demand for copper has not been great. We all know what's been happening in China in the first part of the year. It's picking up now towards the end of the year, that we see more grid spending and there's more spending on the grid, so demand is picking up for demand of copper, but it has not been great during the year. Even if you look at it with the mines that are reducing production and new mines coming up to production, we don't believe there was much supply growth during the year.
In fact, we believe a slight deficit. Even with falling demand, we still had a drawdown of inventories down to 10-day inventory. As you can see from the graph there, that's extremely low if you look at historical years. We believe the commodities which we handle, the key commodities, copper, zinc, nickel, there is a shortage. There is potentially if demand picks up around the world, especially with copper, hopefully next year with more grid spending, more consumption of copper, we will see the market starting to pick up as the inventories are at these low levels. We will need a lot more supply coming into the market.
Looking where we're well-positioned, as I said earlier, with urbanization occurring around the world, especially in China, more urbanization, increase in the population, another 2 billion people in the population by the year 2050, that benefits all our commodities. You can see definitely with infrastructure, consumer spending, et cetera, should be good for all our commodities. Thermal coal, we know we are competing with renewable energy you will have more renewable energy and therefore coal will have to coexist with renewable energy, coal still forms a large base load of the energy supply in the world and electricity generation. To give you an idea, if you look at the energy in the world, we believe today, well we know today coal represents around about 26% of the world's energy supply. By the year 2030, we believe that will go down to around about 22%-23%.
Even with it reducing to those levels, because you've got to remember, renewables cannot fill this gap. Today renewables represents what I think around about 7% of electricity supply. It will potentially grow by the year 2030 to 18%. Therefore coal still has to form a big part of that base load. As I said earlier, going down from 26% of energy production down to 23% of energy coming from thermal coal, you're still going to burn more thermal coal. Today the world consumes around about 7.5 billion tons of thermal coal. In 2030, even with a decrease down to 23%, we'll be consuming around about 7.6 billion tons of thermal coal. Even if you look at just pure electricity, forget just energy, electricity today coal produces around about 38% of the world's electricity. By the year 2030, we predict it will go down to 29%.
Even at that level, electricity today consumes around about 3.5 billion tons of coal. Going down to 29%, we'll still consume around about 3.6 billion tons of coal. Therefore the consumption of coal will continue to increase even with this. Now renewables can fill the gap, but not enough of the gap. We believe coal has a future and will be required for the base load energy in the world going forward. Electrification. We all know about the EVs and the effect of EVs on the major commodities. We're predicting around about 580 million electric vehicles will be existing in the world by the year 2040 and I'll go into more detail what we see going forward by the year 2025.
Naturally that bodes well for above our commodities, nickel, cobalt and copper, which we all know is going to be utilized in vast amount in electric vehicles. Thermal coal, once again I've spoken about it, demand continues and especially in Asia and I'll give you a graph on that later on to talk the effect of the demand for thermal coal and the increase of thermal coal in Asia. Decarbonization of energy, wind, solar, the amount of copper that's going to be required in those traditional energy systems. Renewables will create more use of the copper naturally and as we said the battery systems, storage systems and that's going to have an effect on both nickel and cobalt and so that even as we go more into renewables we'll be utilizing more of these commodities. Once again, I said thermal coal will have to coexist with renewables.
Carbon capture storage I think will become an important part of the energy use in the coal-fired stations. They'll have to have more carbon capture to ensure that we burn coal a lot cleaner. Looking going forward. Oh sorry, wrong one. Just as I spoke earlier about if you talk about CO2 emissions by 2030 will require an additional copper, an extra 22 million tons of copper by the year 2030. That gives you an idea today what the copper market's around about 24 million tons. We'll need to use over the next 10 years another 22 million tons which gives you an idea we will have to grow around about 3.6% annually copper production and as you know during 2000, 2018 we've only grown at about 2.6%.
It gives you an idea just if you look at windmills, solar, et cetera, the amount of copper that's going to be required by the year 2030 with this decarbonization effect, very good for the copper demand. Looking at that if you look at electric vehicles, what is the effect? We've spoken about this before at various conferences and the effect of nickel demand and cobalt demand on electric vehicles. Today the world produces around about, what's it about 2.7 million electric vehicles a year in 100 million car market per annum. We utilize around about 95 metric tons of nickel in today's 2.7 million vehicles. Going forward to 2025 we predict that there'll be about 11.5 million electric vehicles being produced per year. To get there we're going to have to produce another 330,000 tons of nickel per annum to meet that demand.
Today we produce, which is utilized in electric vehicles at the 2.7 round about 95,000 tons. We're going to have to increase 330,000 tons taking it to 425,000 tons of nickel per year produced by the year 2025. Today you know we produce around about 2.4 million tons of nickel. Cobalt is even a bigger story. If you look at cobalt today, as I said, the same story, 2.7 million vehicles being produced today utilizing around about 27,000 tons of cobalt. Going forward to the year 2025, once again, 11.5 million electric vehicles. We're going to need another 73,000 tons of cobalt. That's taking us up to 100,000 tons of cobalt per year. Noting today, we only produce about 120,000 tons of cobalt. There's no doubt the effect of electric vehicles will have an extremely positive effect on the demand for cobalt and nickel going forward.
Coal, as I said before, this gives you an idea of what's happening in the world with coal. As I said, it's still going to be base load. We're still going to be burning more coal by the year 2030 than we do today, even with a decline or percentage of the energy mix. It gives you an idea, if we take a look at all the new coal-fired stations being built around the world today, especially in Asia, by the year 2030, we're going to see demand growth increase by about 160 million tons. Just the graph down below shows you the new capacity which is being built, which areas, Asia, North Asia, + 20 million tons, Southeast Asia + 55, Subcontinent + 55, Middle East + 30.
That gives you an idea of 160 million tons new coal, seaborne, I'm talking about seaborne coal, is required in the world today. The demand does increase. What is happening on supply? That's not stories, that's new coal-fired stations we know are being built and will be up and running by the year 2030. We're talking Bangladesh, Pakistan, India, China, those areas, Malaysia, Philippines, they are being built, so we're going to need that over there. What's happening on supply? We all know supply is getting more difficult to get finance to open up new coal mines is getting difficult. Peter spoke about the permitting issues that you get in various countries to get permits. It's getting tougher. The other thing is Indonesia. Today, we all know Indonesia exports around about 425 million tons of coal, around about 42% of the world's demand.
New coal-fired stations being built in Indonesia. Indonesia is not going to continue exporting 425 million tons, 450 million tons of coal. It will decrease, and therefore we're going to get this supply shortage. There's no question, and that graph shows you over there how drastic the supply shortage can be going forward. If you even look at our operations in Colombia, Prodeco has got a shorter life. Cerrejón doesn't have a massive long life, so therefore you do have depleting coal mines throughout the world. If you have a look at the coal mines in South Africa, we know which ones are depleting and what's happening over there, what's happening in Australia. Not many new coal mines being built.
If you have a look at that graph, if you do have the planned new supply, provided people can get the finance, provided you can get the permits, you still got a large deficit going forward. If people don't build those new mines and there's no expansion, we've got a bunch of mines, Peter gave you an idea of where we could expand our operations, but if we do nothing and we let our mines deplete without that planned new supply, look at the gap we have there. That's the question, what can fill that gap? Renewables aren't going to fill that gap. There's not many other places the world's going to go. That gives you an idea.
Even though coal is so-called a negative commodity, the demand and the power consumption which the world needs, especially the impoverished nations, as I mentioned, the ones where they are building these new coal-fired stations, they're going to need the supply from these operations. As I say, if the planned new supply doesn't come into existence, we're going to have a deficit. That gives you an idea of the different commodities we have and what our priority is going forward. I think we've covered most of these points. Peter gave a detailed description, issues on health and safety and how he's dealing with those, and he's got a big program. He gave the full details of that to ensure that Glencore is fatality-free going forward. Ramping up or the ramp-up development assets. We spoke in detail about them last time.
Peter updated you how they're performing now. Katanga, the one that we had the concern previously, is performing well. Mopani Smelter will start in 2020. Koniambo is the last ones we got to get operating stably and to eventually get it to its 50,000 tons nameplate capacity. How soon, Peter, we'll get it there? Let's wait and see. Operating efficiency and capital discipline. We spoke about capital discipline, how important it is in our company. It's not about just building new mines, increasing production. It's all about what is better to this company, returning cash to shareholders or keeping it in the company to ensure it adds to the value, the NPV of the portfolio, we look at that all the time. Balance sheet we spoke about. Steve gave the full details. We got a robust balance sheet.
We'll maintain a robust balance sheet going forward, and we will live within those criteria which we have laid down for the company. The last thing, we always talk about it, the transition to a new generation of leadership. That will continue to occur in this company. We've said we've gone this, I think I'm the third generation of the leaders of this company. There are a few of us still left, the old guard, and we got to transition to the new management during the year, and we'll do that very shortly. We'll start having the transition moving forward to ensure that we got a top younger, new generation to take over the company going forward.
believe with all those issues, the confidence and the stability of the company is well set for the future, that gives you an idea, making sure all these priorities are put in place. We believe we're in a good position going forward. The company will achieve its required results going forward. Thank you. Questions? Martin, are you doing the questions?
Sergey.
Thank you very much. Sergey Donskoy, Sberbank CIB. Three short questions, if I may. On page 16, the cost guidance, specifically nickel. Looking at this chart, I think that it looks like Koniambo is going to have costs in the ballpark of about 15,000 tons, which taking into account sustaining CapEx means the mine is probably treading water right now. Is this all that can be expected from Koniambo going forward until prices increase and hopefully bring it into a more cash flow positive territory, or you expect some cost reductions down the road? Second question, cobalt shipments at Katanga. I think that from third quarter results, the mine continued to ship below production. When do you expect cobalt shipments to normalize and basically come in line with the production volumes, fourth quarter next year?
Finally, on zinc production profile, you expect production to increase by basically 300,000 tons, to drop by 200,000 tons in 2023, which looks a bit counterintuitive. You're basically increasing production exactly when the market seems to have found a soft patch, which goes a bit against your value over volume strategy. Did you consider maybe taking a more gradual approach to expansions, or was it possible at all? Thank you.
Steve, you can handle some of those. The cobalt, you got those numbers.
Okay. I mean, Koniambo, you're right mathematically that it's about treading water at the moment at the sort of annualizing towards 30,000, let's say, in terms of tonnages at the current pricing of $14,500. That's clearly not where we expect Koniambo to be long term, nor where some of the potential upside may be in nickel margins. There's work to get it. We're annualizing at 30. It's not where we're going to be for the full year this year. We talk about 30-40 as being the range where it's going to be over the next two or three years. At that level, it can start either higher prices or just that increase in production and cost absorption. They are now from having gone through project to an operational like we would do at every asset.
They are looking at where there may be additional cost efficiencies, which is easier to do when you are in a more steady state and sort of rather than a continuous sort of project monitoring mode. Yeah, it's treading water, which is not a bad place relative to where it's been, let's say, over the last two or three years. Once you start seeing that progression towards a long-term 50,000, pick your nickel price, it should start being a cash contributor. Is it going to be the sort of obviously nickel price, whatever it is? It'll contribute nicely over a long period of time because it does have a very long life and extensive reserve base as it has within that business. Anything else on Koniambo, Peter?
I think that covers it. We have a number of initiatives there. Obviously, getting the volume up will make a big difference, and we are looking at cost-cutting opportunities within the business, so working on both the numerator and the denominator to make it more profitable.
Katanga cobalt shipment has been obviously a slow period for cobalt ships. Not that the market necessarily needed more cobalt to be shipped. It just would have affected Katanga's own standalone prospects, which it is obviously as a separate company. Two issues to deal with, one of which was uranium content that was identified about 12 months ago. It's currently processing content that's almost 100% compliant now within the applicable regulations. That's been steadily increasing during 2019. The second issue, which has then been the sort of delay in further generating both saleable as well as executing those sales is the drying capacity, which was delayed anywhere between three and four months, and that's now just up and running. Now we've had the first dryers, Peter said, operating 60% capacity, then another dryer will come on next year.
There'll be a progressive process where it's going to be able to sell what it's currently producing, which is much saleable. If you assume 100% in spec, 60% of what they're producing and increasing over time, that's saleable and will start getting shipped across the border, and there'll be a catch-up over time of the backlog that either needs to get dried as well as uranium dealt with, which there's solutions to do that over time. It'll be a big pickup in cobalt sales, clearly, and will be taking the place of Mutanda in terms of the sort of market-facing volume that's obviously coming through. What's that? On the Zoom.
Just on Katanga, on the cobalt sales, that's really not the end of the world. In the market, that's how we should be supplying it into the market. By the time we do catch up and you're selling your own production against the market, hopefully the market will be in a position that can absorb that, and that's the right place to be. Right now, with a bit of a backlog, not the end of the world suits us.
Just on the zinc volumes, we are seeing, and you would have seen on the graph, that the zinc coming out of Antamina goes up and then reduces again in 2022. That's just a characteristic of the ore body, and we go through this very significant increase. Actually starts kicking in next year for the next two years. There's also a large release of zinc in 2021 from Zhairem project. We're going from zero at Zhairem, ramps up to a very significant amount, then only gets to steady state 2022 to actually drops down again. At that stage, Maleevsky Mine in Kazakhstan is also starting to reduce. There is a bit of a balloon there, but we start leveling out after that at a steady.
Then if you were to say, "Should we be doing it because it's value over volume kind of thing?" It's beyond our control, really, Antamina, because it's a byproduct. Correct. It's just coming. There's nothing you can do about the reserve body. Even if I say to the guys, "Can we reduce the zinc from that operation, et cetera?" it's impossible. You've just hit that reserve body where you've got to do it, produce that amount of zinc.
I think also knowing that we have this profile, you sort of spike up and then you go down in 2022. You have Antamina normalizes, and you have some of the smaller assets in Canada and Peru deplete reserves. This is a production profile. The sales profile could also look different. That's also something you could see some temporary just build up in inventories and sell it over a longer period in time. That's also something that one manages within the business.
Liam?
Liam Fitzpatrick from Deutsche Bank. Two questions on the asset base. The first one, you mentioned the quality at the start, you do have a long tail. Is there any ambition over time to expand the divestment program, or are you pretty happy with the asset base as it sits? Then secondly, perhaps one for Peter. If we look beyond the problem assets at the broader asset base that you have, do you see a lot of scope to improve operations and take out costs, and is there any timing or sort of quantification that you can put around that? Thank you.
Okay. I'll talk about the tail. Yeah, we always look at the tail. We do have a bunch of assets there that really don't add much value, but some of them do help on the trading business, and therefore it's a debate, smelters, et cetera. Which ones do you keep, don't you keep? Does it help on the trading? We're reviewing it all the time. Yeah, I think we will be getting rid of some of the tail assets. Some of it, Steve's got his $1 billion. Some are bigger. Even if you talk some of the listed investments we can look at, and when we do dispose of those, that will be a bigger amount. Some of the tail won't give big value, but it takes a lot of management time away. We continue looking there. That's what Peter keeps looking at.
That's a debate, what value it brings to the trading. Can that market turn? You take, for example, Bolivia. Our Bolivian assets have a good reserve base. There's a great reserve. We never intended developing those assets big because of the political situation in the country. We're carefully monitoring the political situation there. We'll see what happens. If it does become an investment-friendly environment, we could re-look at Bolivia and see what we do there. That's the way when we look at the tail, we've got to look at various ones. There are certain smelters we want to get rid of, don't make economic sense to own, and we continue to look at that. Peter?
Just across the suite of assets in terms of what improvements we can make. The focus that we are having on safety is resulting in some structural changes in those areas where we've got challenges. There are departmental changes, and I actually think that the one will flow into the other. We've got different processes that are in place, and we will see efficiency improvements. A lot of the assets, a big part of the portfolio is going to be managed slightly differently, going forward in terms of operational excellence and making sure that we have reliable, steady performance. That's linked to safety, but it always is also linked to productivity and output and therefore cost. Yes, I do think there are going to be improvements. Those changes are happening, have happened, and continue to happen where they need to happen.
I expect to see that over the next 12 to 18 months, we'll start to see some benefits from that.
Like all the other mining companies, we're looking at technology, et cetera. We got Glencore Technology, the ISASMELT smelting unit, which has a large technology unit. We've got XPS in Canada, which is also looking at different processing methods. We're looking at that. Like all the other mining companies, we're looking at electric machinery, as Peter said. We got the deep electric, fully electro underground in Canada at Sudbury. Like all the other mining companies, Peter's continually looking at that, where we can cut costs with extra technology, et cetera.
Ian?
Hi, guys. It's Ian Rossouw from Barclays. Just a couple of questions from me. First of all, on the operational sort of performance and guidance, just comparing Mopani's guidance to what you said in the August results, it looks like it's already 30%-35% downgrade to the profile. If you can just maybe talk about that. Perhaps just more broadly, obviously you talk about stability and consistency of operational and financial performance in that slide, Ivan. At what stage do you feel you'll have confidence in the operational guidance you give us so that the market can ascribe more, I guess, a higher multiple on these guidance and that we shouldn't expect any further downgrades going forward? Just on the increase in the CapEx guidance, Steve, you haven't really talked much about the Astron business.
Obviously, the returns in the first half didn't look spectacular. It was loss-making. Could you maybe just sort of talk more broadly about that and maybe some of the other investments, sort of higher CapEx and what returns are actually these businesses bringing, given you've, I guess, spent about $1 billion acquiring that business?
Just quickly on the Mopani, you've got two assets sitting in one bar. If you subtract one from the other, you're going to get to perhaps the wrong answer. We have got a little bit of conservatism in that guidance for the African copper assets. My comments about ramping up Katanga, I believe continue to be valid. That's what we're aiming at. Obviously, we are scheduling to restart the furnace at the smelter at Mopani in January. We've added the two together. We are making sure that when you think about the total volumes and then you model us going forward, that we can give you something that we can deliver. There's a bit of conservatism in that.
To your question, when can I have the confidence that we're going to deliver on it all? Remember, we got 150 assets around the world. All our assets are running beautifully. No issues on the assets. Where did we have the issues? You take all our Australian coal mines, South African coal mines, you take our ferrochrome operation, et cetera. Yes, some we've cut back volume out of choice, as I said earlier, but generally, on all our assets, we're hitting our volume predictions. Where we've been missing is these ramp-up assets, and that's what Peter spoke about. Those are the three, Mopani, Katanga, Koniambo. I think if you take those three out of the system, the machine works. Now, Katanga next year will also move out of the system.
It's going to be fully ramped up. As Peter says, by the end of 2020, we will be ramped up to 300,000 tons, copper 32,000 tons, whatever it is, cobalt. That one will also move out. Only other two is Mopani, which has to ramp up. Hopefully, we'll get it done by the end of 2021 and 2020. That's also off the list. Koniambo is work in progress. I think the rest of the assets, comfortable. It's the ramp-up assets we always got to watch. Like any one of the other mining companies, look where anyone who's got a ramp-up asset, how are they performing today? I think we've done pretty well. Yeah, we had a bit of a miss on Katanga.
We didn't hit the 285, and we hit 235, whatever the number is, but not too bad in a second year of production. Next year, we should hit the 300,000 tons annualized towards the end of the year.
Sorry, Ian, in terms of Astron, obviously it's been, what, six months we've been in the chair over there. As you would know, of that investment, yes, there was the acquisition. It also came with a commitment to spend around ZAR 6 billion, which is about $400 million over a five-year period of incremental investment beyond what their steady state was already. We're still looking into both technically and optimizing, sequencing what those might be. We can put more meat on the bone when those finally approve. What we've seen, there's a couple areas that are going to be the focus over there. One, you've got a refinery there that's running about 80% utilization because it's got to go down for a four- or five-week shut every single year. This absorbs the company no end.
It takes six months to plan for this thing, and then they do it over five weeks and no sooner they start to think about the shutdown. That's why you're running. There is some plans as to how you can get that up to about 95%-98% even uptime, and only take it down maybe every two or three years for a few weeks. The other thing, with everything going on in the oil product market, it's also then things around qualities and the types of product that you're actually then producing around the low sulfur, the blends, the IMOs, the fuel, the local things. There's some CapEx. Numbers that we've seen around, those are two of the bigger ideas in those things. These ought to be projects that incrementally deliver returns of at least circa 15%.
Besides that, also what Steve talks about with Astron and the Brazilian asset, they help a lot in the trading, having a short on the crude business, et cetera, also helps in the business, and it benefits on the trading business. Even though we run them independently, we do get the benefit on the trading side.
No, actually, the downstream's working very well, generating good cash flows. The refinery now is the one with this CapEx. As I said, it just feels like it's almost day one still. I think in February, we'll be able to talk a little bit more about it.
Paul.
Thanks very much. Paul Gait, Bernstein. Two questions, if I could. Just on the commodity market, first of all cobalt and then coal. On cobalt, I'm just wondering, have you seen any change in the artisanal supply in Congo over the last 18 months? Obviously, $100,000 a ton sort of incentivized a significant increase as the price has moved down. How is that playing into the supply-demand balance, and what is the fundamental ability of that source of supply to continue going forward? The second thing was on the coal markets, looking at that deficit that you forecast there, and clearly one of the areas where you could see something stepping in to fill that hole would be domestic production of coal. Thinking here, Coal India, also the Chinese domestic production.
Just wondering what your sense is on the price elasticity of domestic supply and how that looks under that kind of scenario? What is the ability of China to potentially increase meaningfully from here, or indeed Coal India to hit the billion ton per annum mark? Thanks very much.
Okay. Thanks, Paul. On artisanals in the DRC, naturally when the price was high, the numbers we got, and naturally it's open for debate how much is really artisanal mining. If we look at the numbers in 2018, artisanal mining was around about 23,000 tons. Today, in 2019, artisanal mining dropped down to about 12,000 tons. You have had a significant effect with the lower prices and that going forward. You saw even the Congolese authorities announce today, somehow they want to put a supply tightening on artisanal mining, that they don't want this extra tonnage coming on the market, and they're going to put some control. I saw some announcement, I don't know exactly where it came, but something that they're going to do there regarding artisanal mining. They've been talking about it for a while. You are correct, it came down.
Now, when you talk about coal, we talk about 160 million tons. Can that be fulfilled by their own supply? Now I spoke about seaborne coal, the 160. So we looking at all the coal- fired stations, it makes clear economic sense only to bring in imported coal. You talk about Coal India. Yeah, Coal India, I think, is producing around about 500, six, seven million tons. Coal India, to meet the demand that is required with the new coal- fired stations being built in India internally, has to go to about 1 billion tons. Whether they get there, that's a debate, how soon they can get there. I think we're talking about 2030, they got to get to 1 billion tons. I don't think they will get there. I think you're going to need even more seaborne coal, potentially.
Where we talk about the 160, and where GAO studies that all the time, he's talking about the coal-fired stations they cannot use or are unlikely to use because it doesn't make economic sense, local coal, and it has to rely on imported coal. We're not looking at all the coal-fired stations being built in India, China, et cetera, which use local coal. We're specifically targeting imported coal. We believe solidly that that seaborne coal, where we talk to 160, cannot be substituted by internal coal.
I guess two sort of related questions. First one's really just on the organization. If we go back to the IPO, the pitch was that Glencore was an organization run by traders and the traders ran the divisions, and we had all the traders up there. Today, we have an industrial guy up there. Is this the new Glencore? I mean, it looks like all those guys are gone and here's Peter, right? I guess a related question, what is the strategy in oil? Again, Alex is gone. It seemed like it was a bit of a failed experiment going into Chad, and now you're putting money back into it.
Good questions. Look, when we did go public also, remember, it was the traders and we had the assets underneath and we weren't a big asset trading company. We had asset managers, Nick, sitting alongside the traders. You remember the model. It wasn't just the traders. The traders weren't running the assets. For example, at the time, there was a zinc guy who ran the assets, there was a copper guy who ran some of the assets, et cetera. It was alongside. When we merged with Xstrata, we clearly did create the trader and the asset manager. Most of the department or all the departments had an asset manager in the asset running the assets. For example, Tor Peterson was the coal trader. Peter was the asset manager for coal. If you talk zinc, you had Daniel Maté, the coal trader.
Alongside him, you had Chris Eskdale now running the zinc. The same for alloys. You had the trader, you had Gary Nagle running the asset. That model didn't change. That model still exists today. Exactly the same. Every one of the departments, you got a trader alongside the asset manager who runs the asset. The trader runs the trading, works very closely with the asset manager to ensure that we're not back to value over volume. He'll tell him what he believes the market can take, what the market can't take. Please, I don't want you just producing for the sake of production. We can't sell it all. Back to the zinc story, back to the cobalt story. That was the trader who said, "Hang on. We don't need all these tons in the market. We're going to screw the market.
We're not going to make money at those mines. It doesn't make sense." That model still exists. What has happened, Peter? Today, of course, you talk $12 billion, $14 billion of EBITDA, $2.2 billion EBIT. Yeah, the trading is an important part of the business to market the tonnage from the asset as best as possible. Third-party trading, very important. That's not going to change. What we've done with Peter now, let's coordinate all those asset managers that they don't operate in silos. That what I felt before, we were operating too much in silos and we weren't getting the message from the top through how we should all run, swapping people around, which Peter's already been able to do. You had me at the top, but I couldn't look into every asset as Peter can because he's just focusing now on the assets.
He's moving people between the different assets. It was just a matter of evolving, finding the right guy respected by the group who could take full control of all the assets with the asset manager still there. Not that much change from the past. Back to oil. Yeah, upstream oil didn't turn out great. It's no secret. We can't sit here and say Chad turned out great. Steve's taken a few write-downs over Chad over the years. Cannot say our upstream adventure was that great. Unfortunately, we did Equatorial Guinea, the first one, and it turned out well. Maybe the success of the first one was the problem of we thought we did have a good understanding of the upstream business. Turned out Chad, it didn't turn out the way we expected. We decided upstream, not for us. We've got a bit left.
As you know, we've got Equatorial Guinea, we've got Chad, and we've got this piece in Cameroon. We continue working that. Will we go more into upstream? No. The downstream business is more of a trading business, you can say. What we like about the downstream business, it's there. It creates a short for the trading business, for the crude, for the products, et cetera, with the distribution networks, which we got in Alesat in Brazil, which we've got in Astron in South Africa. Right now, that's where we're going to stop. We're going to see how it works. Can we get to grips with this business? How much does it help the trading base having the short and just sitting on the downstream business? We'll see how it goes. Upstream, yeah, didn't turn out great.
Myles?
Good, Mr. Chairman. Our oil expert up front.
Yeah, Myles Allsop, UBS. A few quick questions. First of all, on the DRC, could you give us a sense where we are with the mining code and the discussions with the government? With the DRC, how quickly could you restart Mutanda if the market was there f or the volumes? Secondly, on coal, thinking more strategically now. If you continue to trade at a discount and you perceive it's because of being the largest seaborne producer of coal, what options will you consider to try and address that sort of discount?
The third question was going back to the very end of your presentation, when you talked about management and the year of transition. I didn't quite understand the timeframe here. Are you looking at bringing in new management over the next 12 months, or are you here for the next five years, or is it down to the two, three years?
Yeah. Okay, let's take the first one. Mutanda, how quick, Peter, answer that one?
We can turn that on fairly quickly if we needed to. There are some limited oxide reserves, but the long-term future is in the sulfides, and that study is ongoing. I don't think we would. You saw the profile of the cobalt, and you've heard us talk about the sort of growth that we expect in the cobalt market. I think we'd need to be cautious about how quickly we brought that on. I heard your comment, if the market, I don't think we would knee-jerk at a short-term price increase in the cobalt market to bring that back on. I think we take a long-term view in terms of how the sulfide project works and how we ramp it up going through the remaining oxides. There was a question also on the mining code.
Yeah, the mining code, we continue. It is the new code. We had the stabilization agreement on the old mining code, and therefore, we believe that should still be in existence, so we reviewing it, and we continue talking to the government to see what we can do there. We not sitting back. The other one you mentioned, coal?
Coal, yeah.
What do we do with coal? Look, right now, coal is there. It generates a large amount of EBITDA for the company. It's something we believe is good for the company going forward. If anything changes in the future, we've got to review it. Right now, investors have not told us they got to divest from Glencore because of coal. There has been one or just you know, we did do the coal cap. We capped our coal production at 150 million tons, and that's what we've done to satisfy the investors, and that's where we are right now. On management change, I always said, and as I said earlier, we into the third generation, coming to the end of the time of the third generation. There are not many of us old guys left. The old guys will be leaving. How soon? We are reviewing it right now.
I would imagine it would occur next year. We're reviewing it right now. We're looking at it, when the right time for the management change to take place. We're going into the new year. It's time to review it early in the new year, and we will be reviewing it as we get back to work in the new year and decide where we need these last few management changes. It should happen during the year. Myself, I've always said that when the new management is in place, and I hope we can get the new management in place as early as possible in the new year, and then we've got to see once they're in place, I've always said, when one of those managers, there's a lot of rumors who potentially could be there's a good crop of people who should take over.
I've always said I don't want to be an old guy running this company. As soon as those guys are ready to take over, I'll be ready to step aside. It could happen soon. No exact time, but as soon as I believe they're ready, I will move aside.
Hi, Conor Rowley at Credit Suisse. Two questions, so another one on coal and one on Mutanda. On coal, you talked about the evolution of renewables and some domestic coal, but how does the energy market fit into that, given that's probably taken a lot of the sort of price action out of the market this year? Secondly, on Mutanda, you have talked about the future requirements of cobalt, and so presumably the only realistic way to bring that asset back would be the sulfide project. Have you got any provisional sort of indications on how big or expensive that project's going to be?
Peter?
I think it's too early to talk about that project. At the option stage was the pre-feasibility stage.
A lot of things affect for Mutanda, the mining code, effect of the mining code, when to start up, what the cobalt price is looking at. As Peter says.
We have existing infrastructure in place
He's still studying the sulfides. There's still some oxides left.
Yeah.
He's got to review it. No rush. Whilst the cobalt price, no rush. Let's get it right. Let's bring it online when we've sorted out a lot of issues in the DRC and the cobalt price. On coal, I didn't quite understand your question on coal.
Just the LNG market. You've seen a lot of coal to gas switching this year.
I'm wondering how that fits into your sort of outlook on.
Yeah
How much coal you think is going to be needed on the seaborne market?
Debate I'll continue to have with my chairman, but he's the LNG man. We always debate, we talk about this 160 million tons of new coal. We know the coal-fired stations are being built. They're there. They can't switch to LNG, but can LNG plants in those countries beat the coal price? We believe at the coal price, we're talking about hard for LNG to compete. We talk AUD 85, AUD 90 Australian coal, Newcastle type indicator, hard for LNG to compete in those areas. If the coal price goes to 120, 130, yeah, then your LNG could start kicking in. How much, how fast they can replace, we've got to wait and see.
I know in Europe we all said, "Look what happened in Europe." Coal consumption in Europe went down, I can't remember what it was a few years ago, but it's come down to 60 million tonnes in a 1 billion tonne seaborne market. Everyone talks about Europe substitution of LNG gas. Okay, there was a lot of gas also because the gas came, but Europe wanted to get out of coal, went into gas, and they had the gas there, so it's not a big player in the market. The other-- and they're paying more for their energy. The other poorer countries are not prepared to pay for high-price LNG. Even though people may believe it's cleaner or not, that's another debate, but coal is still cheap at those prices.
Thank you.
Agree, Mr. Chairman? See, even the Chairman agrees with that.
Good afternoon. Sylvain Brunet with Exane. Two questions from me left. You talked about the whole EV story. 2020 could be an interesting year.
25.
Yeah.
Oh. I'm 25. 20, I'm not sure of yet. I don't know how many EVs are coming in 2020.
If we talk about the battery chain, what is your assessment of how much inventory is there and what they are doing? My other question is on marketing, which was fairly good guidance. What would be the areas where you've been surprised positively or negatively? Which businesses have done better this year than you would have thought?
Which, sorry? On marketing?
Yeah.
On marketing, simple. We all know oil performed well. You've seen the results come out from, whether you're talking about the oil companies, whether it's BP, Shell, Total, they're all having great results. You look at the other oil trading companies, they come out with various results. Different timing has been good. Our oil trading results are very good. The rest has been okay, and as Steve says, we'll give the number to when the year closes, but oil has performed well. Copper, cobalt, not so great. You know about the situation in the cobalt division, therefore, we had the issue on the cobalt, which is the shifting from the one side to the other side. Overall, oil was a good performer. On the battery supply and the short, I don't know the answer to that really.
Maybe our cobalt guys know that a bit better than me. I haven't seen the number. I know cobalt, where the amount of cobalt sitting, battery inventory, no idea.
I mean, would it be fair to say that our sort of order books and just generally engagement with our customers has been good now on cobalt?
No doubt.
Very diversified, Asia and Europe.
The demand for cobalt has picked up. You saw with the announcement of the shutdown of Mutanda, where the cobalt price picked up, I think, where was it from? It went from about $12-$18.50.
Exactly.
Come back a bit off now down to $16.50, roundabout there. The percentage payables has increased, came down a bit, but definitely increased when we gave the announcement we're shutting Mutanda. Going forward, clear that our prediction is correct, and the amount you read about the amount of electric vehicles that are going to be built in Europe, European car manufacturers, but we see it in the amount of people chasing us for long-term contracts. We've said that we will only sign long-term contracts with a floating price, but people are still locking in tonnages five, 10 years going forward. We've signed a lot of contracts. You see, we keep announcing them as and when we sign them. We've done Umicore, we've done.
GEM
Well, GEM, we've done-
There's various in the pipe, but there's BMW. There's a few others.
BMW we've done. There's a few more coming, and as we sign them, we'll announce them.
That's positive of this battery supply chain.
There's no doubt.
lock things up.
Clean cobalt supply, knowing it comes from an industrial producer like ourselves, most of the companies want to tie up long-term contracts.
Anyone?
Just a few specific questions. One, should we expect some working capital release next year, specifically from the cobalt inventory sales? One could calculate roughly $200 million could come from there if the cobalt that was produced last year could get sold next year. Secondly, I'm surprised that the thermal mine cost guidance hasn't gone up more, given kind of the production reduction that we are seeing, especially in Colombia. What is sort of the break-even volumes there, beyond which kind of more drastic action would need to be taken there? Thirdly, again, on the oil side, going back to the question on Chad, there was news flow that it's up for sale. Is that still the case given the production guidance kind of still looks like there's some growth there?
Are you waiting for all that growth to get realized before you kind of put that up for sale?
Steve?
I mean, cobalt working capital release, yes. We're obviously holding more than what we would regard as sort of normal optimum levels. To the extent we sell down below the levels we are now, then clearly there'll be a working capital release and there's scope over the medium term to release quite a lot there. Whether it's bang smack within a 12-month period for 2020 is not something I can categorically say. There's obviously the trend is going to be towards working capital release within cobalt. Thermal costs, I think, and again, it comes down to the fact that where tonnages come down, that's been the higher cost tonnes that have come down. You have an averaging benefit also from that. We've got slightly better FX rates in Colombia and South Africa on a sort of comparison.
AUD's been hovering at AUD 68, which is obviously the other factor. Various other consumables, procurement benefits. We've had some savings have also been delivered that are obviously baked into some of those numbers. There's also a, not every tonne of coal is the same as the other tonne, and we want to make sure we're taking the most profitable tonnes out. What was the other?
What was your third?
Chad.
Chad. Is it still up for sale?
I think the s-
We look at every asset. Every asset has a potential of being up for sale. At the right price, we sell anything.
I mean, in terms of value maximization, would you look to sell once the volumes are ramped up?
Yeah, we look at its profile. As you see, we've got a profile, and if we get the right price, we look at it.
Obviously, these numbers we've given you assume there's no sale.
No sale.
Obviously the CapEx production cash flows is obviously built in, and if for whatever reason there was something on that front, we would adjust accordingly.
Yeah. Nothing? Thanks very much. Thank you.
Okay. Thanks all.