Katanga at the Glencore Conference call1 A few factors to -- at the EBITDA, a few opportunity cost was the timing of sales versus production both in copper of about 22,000 tonnes but more importantly the cobalt currently that's stockpiling with the excess levels of uranium typically at Katanga. Just in the year 2018 alone we got 3,900 tonnes of cobalt which even at the poor margin -- spot margin environment of cobalt at the end of December had an EBITDA opportunity cost to us of that $134 million. So that's just being stockpiled in anticipation of the ICE plant and the various other treatment initiatives, Katanga also if they haven't already there's going to be their own release before Canadian market goes up, which will clearly go into more chapter and verse around what's going on around cobalt, in particular, around suspension of uranium and how to treat that and how that project is clearly going? So all the costs around 2018 is broadly where we said at the -- towards the end of last year copper is one of four. We guided to one of three at the end of last year and they've come in pretty close to where we said. Increases year-on-year, 2017 to 2018, is clearly being impacted by DRC mining code, higher energy cost and net sales, the production variance which at some point is going to catch up. We'll see the evolution more as we go from 2018 to 2019 as we go across. And cobalt was sort of a double punch, if you like, for us during the, sort of -- from about August, September, October, towards the end of the year in a variety of factors. We've had some Chinese non-performance. We spoke about that as well around certain fixed price realizations on hydroxide. We did have to sit around the table and reset some of those. Hopefully, the swings and roundabouts around that over a longer-term cycle, cobalt is going to have its day in the sun again, but we clearly had to take that performance risk. And also as I flagged within the -- back in December, it's a little bit left-hand right-hand, but within the marketing business we do have to continuously off-take the material from both Katanga when it's selling of course and obviously we've turned around those agreements. So this is coming on to the marketing -- coming on to the Marketing's books. It's not an easily hedgeable commodity. It's not a commodity that you can lay of that this -- the demand strength today. So you will have lag effects of having to take material on to your books and you will have some length in cobalt. To that, we'd like to have less length, but we do have some length in cobalt. And just in the month of December you saw cobalt prices dropped 22% on some index. Clearly, this is not on all the realization. That could turn around as well, but that's what had to be absorbed through the earnings for that particular period. 2 Just a modeling guidance, because that's all looking back, its all going to be more important looking forward. So this is the detail of what comes out of our spot cash flow, it is sort of cash flow generation at spot, across the different Industrial businesses, all the building blocks down the track. Copper is the one that's worth spending a little bit of time. So it's being -- you can see bottom left, I would point towards a range of outcome as probably being something that's more sensible. So weve said $1.4 is the cost. That's what the actual was 2018; 2019 anywhere between $0.92 and $1.25. Now the spot cash flow illustrative $15.8 billion is the higher cost $1.25. So we've taken the most conservative part of that range. What's driving that range $0.92 to $1.25, cobalt. There's nothing else that's driving that range. So at the bottom end of the $0.92, what we've assumed, let's say, in the $1.25 is that we are not selling about 25,000 tonnes of cobalt this year of what we're producing. You can say Katanga. Okay. Katanga but just a general assumption to say Katanga will, of course, it's producing 26,000. They've said they will sell some of their production this year but most of it is going to be 2020. Across there is a modeling assumption we've seen 25,000 tonnes of cobalt not being sold. We've also run through that particular model on spot prices, the realization of cobalt hydroxide as some benchmark would indicate it might be. It's very low volume. It's very illiquid at the moment, but we've taken about $11.50 a pound, is the price thats we want. $11.50 a pound and 25,000 less tonnes of cobalt is what's driving that $1.25 against what would've been a $0.92. That's a $1 billion variance against the cobalt revenue that we're running just two months ago at the end of -- or at the beginning of December, which would have been more sales and higher prices. So you can have your own views as to where the volume and price is going to potentially lead us. And we do speak about the cash flow generation, the $6.8 billion that Ivan said, that's at the conservative $1.25. So I think a range of probably $6.8 billion to closer to $7.5 billion is probably a more bookended around cobalt realizations at the moment in terms of how we go. I'll show you how cobalt, in terms of the assumption, that's the main variable. The other businesses are pretty steady state around its production. And obviously cobalt, there is the assumption then of the -- within copper, we have the expansion 46,000 tonnes this year. That's the net of the extra 150,000 at Katanga and Mutanda being rebased to closer to the 100,000, which we'll see on the slide as well later on. Based on an optimized plan and an optimized cost structure, you would have seen some headlines on that. Where is all the building blocks for all these numbers. That's the production longer-term guidance unchanged in cobalt and all the other metals. Copper is the only thing which is a drop of 40,000 across all years on account of Mutanda. Even 2018 to 2019, we've got some reasonable volume growth in cobalt still sort of query the sales impact on that, but physically you'll be able to go and see the bags of hydroxide there. And ultimately it will convert in to cash. Whether it's a 2019 or a 2020 story, we'll obviously wait and see 3 Cobalt, we are -- obviously as and when the industrial assets produce they will sell almost instantaneously to Glencore. And this is -- we have metal and we have hydroxide. The hydroxides coming out of the African assets, Mutanda that's about 50-odd-thousand probably between the 2. And then although 57th there's another 5000, 6000 more in metal coming out of Murrin, out of the nickel business Murrin in sort of iron ores. So that's a more of a easier commodity if you look at the sort of metal and how that -- and that's slightly more hedgeable and there's a more functioning market, the hydroxide market is maybe a bit more complicated at the moment. So obviously our assets are exposed to the spot price to some sort of realization of what price so the benchmark times the payability and we've assumed year-end $11.50. So I mean that's been significantly higher. It's been a touch lower as well. And that's -- that market that we think is starting to sort of bottom and find its strength and has displaced some of the higher cost production. There is a lot of swing production I think as I mentioned earlier particularly some of the artisanals, some of the handpicking production that may come out of DRC. When prices go higher, so obviously you see a hub of activity and the buying continues prices are down here and clearly less of that going on at these sort of prices. Within the -- because of a period of oversupply within the marketing part of business, we can't tell Katanga or -- that sorry we're not buying you, the market's soft we're, not buying you we need to continue to take it. Now, obviously, more difficult to hedge clearly in that commodity, almost impossible I would say. So, at some point, you will sort of have to buy it. And then you have some -- you are carrying cobalt inventory at a certain price. And if prices go down, you have to take some NRV provisions against that. Now, we're trying to keep that as low as possible. And we have a big access to the customer base and we're obviously very involved and very strong in that market. But there has been some buildup in being long cobalt; we are fully long cobalt at the moment within the marketing part of the business. Not to any point that's -- it's going to be materiality that it's going to -- that needs to be flagged unusually, but that has been a factor and will continue to be a factor. We see the ability clearly during this period that that ought to be stable to sort of going down, particularly in fact, Katanga is currently not selling and there will be a period. So, absence Katanga being in the marketplace at the moment is still very healthy demand growth and where prices now pushed down the -- some displacements of some costs. This market is starting to sort of feel a bit more functional and we should have a clearer pathway in terms of cobalt. 4 Jason Fairclough So, you guys are naturally long because you've got the flow coming through Katanga, but ultimately, you can't sell that Katanga material at the moment. Or are you taking delivery of that Katanga radioactive stuff? Steve Kalmin No, no. No, we can't. Ivan Glasenberg No, we don't take delivery of it once it's radioactive. We will only take it when they clean out the uranium. 5 Jason Fairclough And does Katanga have it on at cost or is it been mark-to-market or--? Steve Kalmin No, Katanga has it on its books at a low cost of production; it's sort of the by-product accounting that goes on. You'll see in -- you'll see even that -- I mean we spoke about this 3,700 tonnes or 3,900 tonnes which is what's been produced and unsold at the end of their books. Even at the low prices on some mark-to-market at the end of December, had they sold that at that point, there was an opportunity EBITDA loss of $134 million. So, there's a lot of buffer at the Katanga level relative to what they carry. And they're carrying at very low levels of production costs at the by-product level. Ivan Glasenberg It's way below net realizable. Steve Kalmin $134 million. Ivan Glasenberg Below net realizable. Steve Kalmin At the end of December. 6 On cobalt, you talk about the market and the consumers how they're behaving. Look they did stock up when the price was running up. We know that was obviously what drove up the prices, the battery makers were holding up inventory. You also had -- with the higher prices you had as Steve mentioned earlier the artisanal mining, they'll supply more during the higher prices has not come back. And if there's artisanal material coming onto the market. The buyers you've also had the new tonnage will be coming from the ERG operations in the DRC. So that is adding new supply. So in 2019 you do have a lot more supply we are -- okay subject to the Katanga issues and the uranium, but we also added a lot of new supply. So the market came off with this new supply coming in the market. As we said earlier it sort of seems like it's bottoming out. People are destocking in China and therefore they will have to come in the market and start restocking. And as they see the price move up you will see that happen. So yes, we're watching. 2020 we definitely believe the demand will be there, more demand, more battery, more electric vehicles etcetera and supply may not have caught up by then, but we got this balance in the middle in 2019 and let's see how it washes out. But it is showing signs of artisanal having cut back, us naturally with Katanga tonnage all coming to the market is starting to put a bit more pressure on pricing. But we'll see where we go in 2019. Ivan Glasenberg We see a lot of consumers are coming to us to lock in long-term supply. The tonnage as they talk about going forward in 2020, 2021, 2022 are really significant which they want to lock in with us. Now naturally we don't mind locking in tonnage subject to how we price and etcetera so we're looking at various opportunities. But clearly, the battery makers do see cobalt being an important metal and seeing that it's not readily available, it is not large tonnages going forward because the increases we talked about that occurring in 2019 with us and the Kazakhs etcetera, we post 2019 don't have much increase. The Kazakhs post 2019 don't have much increase. So there's no -- at the end who's going to be the balancing at the artisanals and that all depends on price. So if we go back to the higher prices, whatever it is $80,000 again yes you may get more artisanal, but also artisanal is limited. So the buyers are seeing this and the buyers are definitely trying to... Steve Kalmin Lock it up. Ivan Glasenberg Lock up tonnage supply. So 2019 we got this blip. 2020, we believe it starts moving again and therefore -- and thereafter there's no new supply with demand continuing to grow. There was a third question?