Ivan Glasenberg - Chief Executive Officer and Executive Director Steve Kalmin - Chief Financial Officer

Ian Rossouw - Barclays Jason Fairclough - Bank of America Sylvain Brunet - Exane BNP Paribas Sergey Donskoy - Societe Generale Myles Allsop - UBS Bank Menno Sanderse - Morgan Stanley Tyler Broda - RBC Capital Markets Liam Fitzpatrick - Deutsche Bank Dominic O'Kane - J. P. Morgan

Good morning. Thank you for attending the call. I'm sitting here with Steve, the CFO. And we'll take you through the financial highlights on the first half of the year. As you'll see from the presentation on Page 4, the highlights, we had a very strong financial performance during the first half of the year. Adjusted EBITDA is $6.7 billion, which is up 68% from first half last year and adjusted EBIT is $3.8 billion, up 334%. Net income attributable to equity holders is $2.5 billion as opposed to the first half last year was negative $369 million with write offs we took during that period. The funds from operations is up at $5.2 billion, which is up 88% from the same period last year. Naturally, these results are underpinned by stronger commodity prices and higher margins, which we've achieved based on the higher commodity prices period-on-period. And as you can see, if you look at the various commodity prices, copper is up 22%; cobalt, up 109%; zinc, up 49%; and thermal coal realizations, up 50% to 70% on -- within the same period last year. The company once again has proven that the reduction of the cost structures across the board of our assets has been maintained during 2017, and we continue pushing the cost structures across these Tier 1 assets. And as you will see, our costs are extremely low across the range of the commodities. Copper, where we're producing copper at $0.88 per pound; zinc at negative $0.09 per pound, if we take the gold credits against it; and nickel at $2.40 per pound; and thermal coal, $45, which is giving us significant margins roundabout $32. So once again proves the strength of our asset base and the quality of our assets, whereby we are able to produce at these extremely low costs. The Marketing business, which we always say is a good part of the Glencore business. It's resilient and it's not linear to commodity prices. Naturally with higher commodity prices, it does perform better, but as Steve will show in the slide later on, even with the lower prices since we're public, it's been extremely resilient. And you will see the Marketing adjusted EBIT is $1.4 billion in the first half of the year. That's up 13% on the same period of last year. But if we do like with the like now that we only account for 50% of the agricultural business, is up 22% on last year. Reflecting a solid year-to-date performance, we've increased the guidance for 2017 by another $100 million. And therefore, we're indicating guidance on the Marketing business EBIT, $2.4 billion to $2.7 billion. Naturally, with these stronger cash flows and stronger results, the net funding and net debt has reduced by $2.4 billion and $1.6 billion, respectively, over the first half to $30.2 billion and $13.9 billion net debt at the end of the half year. We are -- robust cash flows are being generated. And therefore, the cash flow ratios are very strong. FFO to net debt is 74%. And you can see the net debt to adjusted EBITDA is down at 1.07x. This is the lowest level the company's had in its history. Turning to Page 5, sustainability and governance. As you will see, unfortunately on safety, we've had 4 fatalities from 4 incidents across the range, but we've had a substantial improvement in this area of our business. And as you will see from the LTIFRs, we've reduced 64% since 2010. But unfortunately these fatalities, three of them are at our focus assets. As you know, we work in particular areas where it's difficult to change the culture. We've been working strongly in these particular areas, but three of the four incidents did happen in Kazakhstan and Bolivia. And these are part of our focus assets where we are focusing to reduce the fatalities, and a large amount of the employees are employed in those areas. And we'll continue work aggressively to reduce the fatalities and increase the safety performance in those areas. But noting we do employ 150,000 employees and contractors across the range of our businesses. So with that, I'll hand over to Steve, and he'll take you through the detailed financial performance of the company.

Thank you, Ivan, and good morning or good evening to everyone, depending where you're dialing in from. If we just turn to Page 7, financial highlights under four or five different headings there. There is more details later on in the presentation, so I'll largely skip through this, but it just shows the strong financial performance; the low-cost, high-margin industrial asset performance and portfolio that we have; embedded cost savings that are now coming through. And we'll show the delivery of our cost structures, cash cost relative to guidance and as well as for the full year. We'll get to slides later on. A huge strengthening in capital structure. The graphs are now turning very positive. Again, we'll look at charts later on, on that. And as Ivan mentioned, the lowest and strongest financial coverage ratio this company has ever had in its history. We'll reinforce the work through the capital allocation chart that we went through towards the end of last year and how that's been applied during the course of 2017. And what we intend to do now during from six months -- to six months is look at resetting and going through the illustrative annualized cash flow generation of business based on spot prices, which -- we'd said this three, four days ago, prices have in fact ticked up a bit. And if we fully mark-to-market, that would be comfortably above the 15 billion. But we've set out the main business units in coal, nickel, zinc and copper at 12 billion EBITDA spot prices. Some of the smaller divisions, the Marketing EBITDA, that's the midpoint of guidance at the moment at 2.7 billion. The midpoint of our long-term guidance range would even be higher than that. That's where the 15 billion comes from, and the cash tax and interest at that level is 3.8 billion with interest at about 1 billion. That's even at slightly higher interest rates, so we've built in potentially one or two other rate rises at spot just to build through the interest. Taxes would be 2.6 billion. That's before any usage of some tax losses which we still have carried forward within various jurisdictions, and we're obviously working through those quite quickly at the moment. But the actual cash generation will be more positive than this because that's a tax base unhinged from any capital losses or tax losses. That's 2.6 billion. We also put 0.2 billion there, so 200 million that we take out of EBITDA for some noncash that's in the EBITDA that we're working through that number. And that's essentially made up of some of the streaming transactions for gold and silver that's being amortized back in out of the deferred revenue. CapEx is the 4.1 billion to give 7.1 billion. Now all the details of how those numbers are constructed, we put two or three pages out in the appendix, pages 21 through 23, which shows the cash flows commodity-by-commodity based on our production guidance, the cost guidance which we look at for 2017, and we work through all those numbers. So that was $4.8 billion for copper, $2.6 billion for zinc, $0.5 billion for nickel, $4.1 billion for coal. And you can see the buildup getting to the $15 billion. It's worth noting just for coal at the moment, because it's a commodity that's maybe less influenced by the -- as in metals, by the spot price from time to time. There is a collection of longer-term fixed-price contracts. There'll be quality discounts. You've also got a curve currently that's reasonably backwardated. We've been quite conservative in the projected margin within our coal business. You can see we've got Page 22. We put a margin of $31 a tonne, which is based on -- off the Newcastle, more of a forward curve scenario looking into Cal-18 of around $83 off the Newcastle with an adjustment of $6. If we had to really use the spot price today for Newcastle in coal, which is very strong in the sort of mid-90s, you would add -- if that was to stay flat, there is a backwardated curve, the coal business would increase by well over $1 billion in terms of a typical spot analysis. But we are being conservative there, reflecting the backwardated curve and reflecting some of the quality adjustments and the timing [does]. So with all the commodities, coal is more looking at where we see the full year 2017 rolling out in terms of the coal earnings at about $4.1 billion. So we have -- the appendix is there, and I'd certainly encourage you to work through those. If we go through to Page 8. Now just working through the more details. It's the Marketing adjusted EBIT. On a reported basis, up 13% to $1.4 billion. I think as Ivan mentioned, that's actually up 22% with agricultural on a like-for-like basis. As you would recall, we sold the business, 50% of the business which closed in December last year, from which we moved to a proportionate 50% reported consolidation basis going forward. So the comparative numbers in the agricultural, the $115 million, was on 100% basis in H1 2016. We got $107 million on a 50% basis. So if you were to report that or add that back at 100%, you would see the plus 22% and the strong performance from the agricultural side. Good contribution, some good increases across all 3 segments: 23% in the metals; 15% in energy; and agricultural, as we said, up 86% with $1.4 billion, with generally supportive market fundamentals and conditions in each market. That's a good tracking year-to-date's performance. If we annualize that, we'd be obviously above $2.7 billion. So that would be tracking towards the upper end of our guidance range of $2.4 billion to $2.7 billion. There's nothing that we're seeing in tea leaves in terms of second half performance. The second half has started well as we've seen in January and a little bit into August. So we would certainly hope and expect to finish towards the upper end. But in terms of just maintaining that range through a conservative bias, we'd look towards the Q3 period when we report production in October 2017 as another period in which to maybe narrow guidance and hopefully, even be able to uptick that guidance a bit on the Marketing side as we look forward once we've got through Q3 2017. If we go through to Page 9, you can see a longer-term historical reporting of the Marketing business and where it's come within that range. The tick-up in range back in '13 was with the acquisition of Xstrata and the synergies that came with that. And then the slight tick down is by virtue of having gone from 100% to 50% on the agricultural side. I think the bottom-left graph is quite powerful, picture telling a thousand words. You can see that resilience and greater stability in the Marketing part of the business relative to commodity prices. So you can see that on an index basis compared to how you've seen the, obviously, contraction and then pick-up within the Industrial part of the business on the lower graph. So we've annualized -- for 2017, you can see in the Marketing part, that's going to be stable relative to 2016. On a like-for-like basis, that's the dotted line with 100% of agricultural, you would have seen an uptick. So there would generally be some positive correlation towards higher prices, but it's much more both resilient and stable, less correlated, but supportive part of the business that generates a lot of free cash flow off the fact it has low interest rates, it has low tax base, it has very little maintenance capital that goes into that business. And at the bottom right, we talk about that being a very resilient, high-return-on-equity business as well. So that's the Marketing part of the business. On the Industrial part, if we go over to Page 10 and some of the increases and waterfalls that we provide from period-to-period. So that's up 95% to $5.3 billion, with strong increases both on the energy side and the metal side. The margins itself, EBITDA margins across the mining side of metals, you can see on the right side, that's up from 28% to 38%. I'm sure you'll see that plot extremely favorable against peer analysis in terms of margin. That reflects the low cost structure and the robustness of that particular business. Similarly, the coal mining margins moved up from 17% to 41%. If we look at some of the variances as to going $2.7 billion up to $5.3 billion, of course, pricing was the major impact there, contributing a $3.5 billion positive variance. That's made up about half and half between the metals and the energy, $1.75 billion, $1.75 billion, making up the $3.5 billion. If we break that through on the coal side, $1.75 billion, that's with realizations on the thermal coals. As you'd be aware, we have exposure to both Atlantic and Pacific. We've got Colombian origin, we've got South African, we've got Newcastle origin, different quality brands. But our thermal coal realizations, even including domestic, were up between 50% and 70%. And our coking coal realizations, that's semi-soft and hard coking, were moving double from the corresponding period. So that was up $1.75 billion. On the metal side, $1.75 billion as well; $0.8 billion coming through the copper industrial business, which is supported by both copper and cobalt, itself, is a significant by-product. Cobalt, up 109%; copper, up 22%; zinc was up $0.5 billion on the basis of a zinc price up 49%; $0.3 billion increase in the ferroalloys business with high realizations on the chrome business, and a little bit on the other side of that business. So the key variance, as you would expect, was price. There's been some minor cost, FX and inflation adjustments the other way within the volume variance. A little bit in the copper business, where there's some variability in end-of-life volumes and cost related to Alumbrera, which has got a short mine life left now; so that had a volume variance as well, that 17,000 tonne reduction or 45% from the first half 2016. We've noted also some lower African copper volumes, 19,000 tonnes, primarily weather related there to do with Mutanda, and a smaller effect as well. You'd recall, we sold 30% of Ernest Henry business in November last year. That also took out about 7,000 tonnes of copper, which would've contributed somewhat there. And some of the oilfield depletion as well, which would've contributed a volume variance in terms of contribution there. On the cost side, we've seen some higher cost there to do with grades at Antapaccay, mostly temporary there. That should pick up as we move forward. The coal portfolio mix as well. It's -- it doesn't come at a margin. But sometimes, you'll have higher cost depending on your mix of coking and thermal. And expansions as well as increases in some JVs that we had at Newlands, Collinsville, which we acquired the 45% of those assets we didn't own last year. And there's been a slight uptick in some of the more passive energy-related costs that have come through. You have seen Brent up 29%. We're a big fuel consumer in the diesel side of our mining operations as well. And a little bit in Kazzinc, which has a mix of hydro and normal grid power. There was less hydro draw, which is a much cheaper part, so there was a bit of cost variance coming through there. On the FX side, there was $172 million variance there. That's with the South African rand having been surprisingly strong during the first half. It's since weakened with developments in the last few days. So that should go positive in the second half of the year. And the Kazakh tenge, the Australian dollar and Colombian peso were also currencies that appreciated during the period. If we move through to Page 11, you can see the trajectory in our unit costs of our -- some of our main mining businesses and some of our commodities. Now this is -- as we put the reconciliations through the back of the pages on 21, we've given you the half -- the first half reconciliation between production guidance, between costs, actual EBITDA, reported and what the guidance said at the time. So there's full transparency around these. So actual cash cost that can derive the fully loaded built up EBITDA cash flow generation. So it's including all the mining cost, all the freight cost, the treatment charges, the smelting and the entire cost of the business as opposed to some of you might be more familiar with the sort of the C1 cost definitions. This is much more fully loaded in terms of building up the cash cost. Within the copper business, you can see, 6 months ago in February, we gave guidance of $0.89. For the first half, we've come in at $0.88. And for the full year, we're now dropping down to $0.86. Some of that clearly helped by the cobalt by-products given the price elevation there. If you looked out at the nickel business, we guided towards $2.48, that's come in at $2.40. So good cost control generally, despite some FX appreciation on the -- or weakness on the dollar and some fuel related increases and a little bit -- and some volume contraction. Again, nickel business enjoys by product credits through copper and cobalt coming out of both Canada and Australia. For the full year, we're expecting cost to go down to $2.13. The zinc business has maintained a cost structure of about minus $0.10, minus $0.09, both in the guidance full year as well as half year. And that's with a slight tick up in cost at the zinc level but strong gold performance, production performance out of Kazzinc. You'll see if you go through the numbers, there was a 26% increase in gold production out of that particular unit and an extra 60,000 ounces of gold, which at fairly stable gold prices was a very strong performance in the zinc business. And the coal business, a very pleasing overall cost performance there, $44. You can see $45 actual, $46 guidance. You have had the more potential headline pressure coming in costs at the coal business, given in our various structures you have royalties linked to prices, which have been very strong. You've got fuel related costs, and you've had currency. So strengthening Australian dollar, South African as well as Colombian peso. So they've been able to hold costs fairly well with some productivity improvements. We highlighted about 300 million targeted cost reductions through the coal business at the end of last year. That's about a two year project. And that's being delivered through the system, and that's the margin of $28 was guided. We've come in at $32, $31 or so for the full year as well. But in a true spot scenario, that would be much higher given the strength in coal prices that we're seeing at the moment. If we go to Page 12, just an update on CapEx. You can see year-to-date CapEx, 1.6 billion, tracking low on an annualized basis to the 4 billion guidance which we have for the year and we've consistently maintained throughout the last 12 months or so. That's made up on a full year basis roughly at 3 billion sustaining, 1 billion expansionary capital, and expansionary coming primarily through the projects that people on this call will be well aware of. That's the African copper, Katanga Whole Ore Leach. That's one of the big projects. Mopani is sinking three shafts as well as ultimately putting in an updated concentrator. And the rebuild of the Line one and two at Koniambo, and the capitalization of costs at Koniambo until it reaches commercial production, hopefully sometime in 2018. So we're tracking below the annualized in terms of the bottom-up reporting through from sites. There's still an expectation, and there's a physical sort of activity set within the timing of 2017 that would calculate up to about 4 billion. Of course, timing wise, CapEx is a bit more difficult to estimate exactly when the cash flows associated with these will occur. So there's some prospect that some of that cash flow could move from '17 into '18, but for now, we're going to hold the 4 billion. And that should be a level at which the company can function at least over the next three to five years, including ultimately that would be the delivery of some of these expansionary projects, and certainly allow for production increases in the optionality within copper and zinc at some particular point in time. This business, as you can see, the reason for maintaining the graph on the left is just to highlight how much capital has been spent over the last 10 years, more than 38 billion expansionary, 66 billion, so the entire fleet in the Industrial is very well capitalized. It's fairly modern, and it does sort of produce, say, a fairly low level of sustaining CapEx for the level of activity we have for the foreseeable future. If we move to Page 13, you can see the balance sheet further strengthened into very strong territory as we would see it at the moment. So you've seen both net debt and net funding reductions to 30.2 billion and 13.9 billion, respectively. Those charts on the right are quite telling of the journey through peak debt around 2014, with CapEx coming down as well as some debt reduction plans that we put in place through the last two years or so. So the FFO to net debt, net debt to adjusted EBITDA, some of where we're going to is almost off the chart on some of those ratios that we can see. The long-term movement, as you recall during the last 12 months, was a repositioning or a recommitment to maintain leverage within the business of no more than two times net debt to EBITDA. Previously, that level was set at three times. That, we felt, was consistent with and appropriate for the business to introduce greater consistency and to reset a more conservative profile. We moved it through to 2 points two times. If you look historically on the bottom right, even at three times ratio where we were managing business historically, it was never breached during this period despite the sort of commodity upheaval that we saw in the last sort of 12 months or so. So with EBITDA sort of $15 billion, notwithstanding the repositioning and how we were tracking the business, this business was -- still continued to generate cash and had good coverage even throughout those periods as well. We've introduced -- which was always something more, I would say, that the business was informally and working towards, was both to have it two times but also to have a sort of a cap as sort of a debt ceiling, if you like, above which we would maintain absolute levels of greater conservatives than we've historically had. That's where you can see the manage around the net debt cap of $16 billion, which was the opening debt that we came into this year of 2017, which is $5.5 billion. So unfortunately, that kills my CEO's comments earlier in the year where he hinted towards the $20 billion dividend at the time. But getting down to, I think, $10 billion to $15 billion, I think is a sensible range. Anywhere below $10 billion of which we're moving very quickly with the sort of cash generation towards, that would then allow us -- there's a slide on distributions later on, that would certainly allow us to materially increase that minimum payout ratios to the point, if there's no other opportunities in reinvestment and some growth initiatives, where you could see the payout ratio moving sort of towards the 100% once we're below the $10 billion in net debt, which sort of confirms that sort of range. If we move to Page 14, you can see the capital allocation wheel, which we introduced December last year and covers the distribution policy that was introduced and how we're applying that both through '16 and into 2017 and '18. Maybe if we just go through the -- at the top right, the green part of the circle and the priority, clearly, to maintain a strong BBB, Baa, which has been signaled and has been a priority throughout the business. We are at BBB flat today with positive outlook from S&P and Baa2 from Moody's. We think the trajectory and the strength of the business is both consistent with financial policy, hopefully moving up towards the upper band of that. So we started the year -- you can look in the text that we started the year at $15.5 billion, with a comfortable 1.5 times net debt to EBITDA, 50% FFO to net debt. If we continue following the arrows across the equity cash flows during the year that had been generated, it would've been $3.2 billion. That's $6.7 billion EBITDA, minus $0.6 billion tax, $0.8 billion net interest. Net interest is much higher in the first half in cash terms than the second half given coupon payments, so that should even come down in second half. $1.6 billion was the net CapEx. And net-net, this is non-RMI, because this is at the net debt. There was actually a net increase in the working capital via receivables, payables of about $0.5 billion. That includes a little bit of distribution to minorities as well, primarily in Kazzinc. If we follow that through, so that generated strong cash flow there. We paid out the first of our $1 billion distribution, $0.5 billion back in May. The second installment will be paid next year. I would see the blue and the sort of brown tranches there, part of the pie, almost working as a 3A and a 3B. Because depending what you do in the M&A and the other, it obviously allows you to make bigger or smaller distributions as well as to -- and also judge how you want to end the net debt. In terms of any acquisitions or M&A activity, back in February, we purchased the minorities out of our African copper business as well. So that was Mutanda. We moved from 69% up to 100%. And in Katanga, we bought about 10% minorities here to move from 75% to 85% of that. Within the second half of the year, there actually would be some disposals coming through as well. We've announced two transactions where there is cash due. One of them is a transaction that should close fairly soon, which is selling our two African zinc businesses to Trevali for a mix of cash and shares, a $400 million transaction: $200 million cash and $200 million equity. That will obviously be proceeds coming in. And our oil logistics terminaling part of the distribution business there. There's a 51% sale to HNA Group for $775 million, with the expected closing also in the second half of 2017. So this -- I mean, that's where there's been some positive cash clearly generated the last two years through asset disposal, which allows both debt reduction and distributions to be paid as well. There was also this year a 0.5% FX revaluation movement associated with net debt, given some of our bonds are in dollars or sterling or pounds. There's been a general weakening of the dollar, which has seen a revaluation of the non-dollar debt, which is all hedged. We've swapped that back into dollars as well, but you can have a sort of a lag effect, if you like, between the translation and the sort of collateral margining of some of those transactions. That particular gap of that $0.5 billion, it's in respect of primarily almost exclusively to do with some old Xstrata bonds. Those are -- that difference on the FX movements to do with non-debt is going to contract in the next year or two towards zero. There was just some timing mismatches historically. If we look at the distributions then, so we'll pay the second half of the -- second $500 million of the $1 billion we said we'd pay in 2017. And as we look towards 2018, when we report results in February 2018, we'll look at the actual cash flows in 2017, and we'll look to implement and apply the distribution policy in its first fullness. It will be made up of the $1 billion fixed plus minimum 25% of the industrial free cash flows, which is EBITDA less interest and tax allocation and the CapEx that's paid within that particular business. If we just look towards our spot annualized cash flow currently that we've highlighted, the $15 billion EBITDA, $7.1 billion free cash flow, applying that formula would produce a minimum distribution of about $2.5 billion, which would be the minimum, then we'd look on an annualized basis to then declare a base of this policy, which represents about a 35% payout. Of course, that can be fully flexed up from their basis with net debt, what the prevailing market conditions and environment are and the context of how much surplus capital as we sort of get our net debt comfortably below that $10 billion. Notwithstanding how much we may declare in February, let's assume it's sort of around that sort of level. We'd also have the ability at interim reporting ongoing. So if we get to, say, August 2018, even though we would have been paying, say, $2.5 billion, 1/2, 1/2, $1.25 billion in May, $1.25 billion in September, at each interim reporting, we would have another opportunity to look at the overall macros, look at market conditions, look at actual cash generation, look at M&A and have the ability to top up distributions as appropriate. So that would be part of our interim review as well. So with that, I'll hand back to Ivan just to close out the presentation.

Yes. Thanks, Steve. If you look at Slide 16, I think that's an important slide. We try to emphasize that we believe Glencore has what we call Tier 1 commodities besides talk about Tier 1 assets, where we have Tier 1 assets where we produce low cost producing commodities. But also, we believe we've got Tier 1 commodities if we look what's happening with the world as we move towards more maturing economies. And the early cycle commodities that underpinned the supercycle boom was mainly from fixed asset investments, and these are likely to be replaced as economies or GDP per capita move in various countries towards higher earning basis. As we are aware, today, China has a GDP per capita of round about $8,000 to $9,000. And as I move along the curve and with -- towards the United States or where we got around about $60,000, you go more towards early cycle, late cycle commodities. So what we talk about, you'll see on the slide on the right, we talk about early cycle commodities is naturally iron ore, coking coal and manganese; mid cycle commodities, copper, zinc, nickel, aluminum and lead; and late cycle, cobalt, oil, gas, thermal coal and agricultural products. So we believe Glencore is well set up with these late- to early-cycle commodities, what we call Tier 1 commodities, moving towards coal, agricultural products, copper, zinc, etcetera. So that bodes well for us in the future. And we should see strengthening demand for those commodities as economies grow towards later cycle. So if you look at the next slide, we have a lot of supportive fundamentals for our key commodities. And if you look at what's happening in the world today, it's the most synchronized global economic growth environment we've seen over the last six years. There's solid demand for the commodities as we've seen by the increase in the commodity prices over the last year period, which is boding well. But what is also good is we haven't seen a supply reaction from the mining companies. As you are aware with the tough times that we've all been through since the 2015 drop in commodity prices, not many of the producers have big supply projects coming forward. So what we see with this demand growth and with the PMIs growing in all these different zones across the world where everyone is synchronized, we will need more demand for commodities, but the supply growth is being limited and Greenfields are not on the drawing boards of many of the mining companies and there's limited Brownfield opportunities to increase production. So this definitely bodes well for commodity prices going forward. We talk about Glencore, which is good for us. We have the greening of the global economy, which is underpinned by policies to curb greenhouse gas emissions and air pollution. And this is definitely being led by the electric vehicle phenomenon which is occurring today. We've done a slide to show the amount of nickel, copper and cobalt which is utilized in electric vehicles going forward and the amount that has to be utilized: copper, 50 to 75 kilograms; nickel, 40 to 50 kilograms; and cobalt, 5 to 15 kilograms. That displays what is being utilized in the batteries. So if you take electric vehicles today, the amount of cars which are produced today, today we only have about 2 million electric vehicles being produced in 2016, but you can take various different scenarios that people talk about, trying to have 30% electric vehicles by the year 2030. It -- just looking at those figures, they're phenomenal because today you produce 87 million cars a year. That would indicate around about 26 million electric vehicles being produced in the year 2030. So to give you an idea, that would require annually round about, if you don't even take the charging points and the requirement of copper for the charging points, just the battery alone, you're talking about another 2 million tonnes of copper being produced per annum. You talk around about 1.2 million tonnes of nickel being produced per annum. And we all know today, we only produce around about 2.1 million tonnes of nickel per annum. So really, if this does occur, you need another 1.2 million tonnes. The same if you look at cobalt today, you'll require around about, at that amount of electric vehicles being produced, 26 million vehicles, you'd need around about an extra 260,000 tonnes of cobalt per annum. Noting today, the world only produces 100,000 tonnes of cobalt. So this is a phenomenon that, whether we believe 30% electric vehicles are up, but even if you take it lower, 20%, and people have different scenarios out there, it definitely will have a major effect on nickel, copper and cobalt prices and the amount of production that the world is going to need. So looking at how Glencore, if you look at Slide 18, we believe with that scenario, we're well positioned for the future. What we like about our portfolio, if you have a look at the slides, our earnings are very diversified by country. We are not reliant on one particular country. We're well structured around the world and diversified around the world. We have a great diversification of our commodity mix, as you will see on the slides there, zinc, oil, ferroalloys, nickel and copper. We're not dependent upon one commodity. Looking further, as I said, we have Tier 1 assets, and that is well demonstrated by our cost profile. Steve has taken you through the cost profile of our commodities. Copper, where we're producing at $0.86 per pound. Zinc where we are, nickel where we are, very low cost producer of thermal coal, where we have relatively high margins at current commodity prices. So we're really well positioned with our asset base, long-term asset reserves and very low cost producer. We talk about the right mix of commodities, which I've just mentioned earlier. We believe we have the right mix. Copper, cobalt, nickel, very important in the areas of electric vehicle, but also the late-cycle-type commodities. If you look at, going forward, the growth prospects the company has without entering into Greenfield expansions. We have our copper profile, where we talk about the extra 400,000 tonnes we'll be producing over the next two years. We have Katanga, which will come in operation during 2018, 150,000 tonnes of copper in '18 plus another 150,000 tonnes coming from Katanga in '19, being a 300,000-tonne producer, which will also produce roundabout 20,000 tonnes of cobalt when it's in full production. And then we have Mopani, which these shafts, which Steve spoke about, and the concentrator will be completed at the end of 2018. And that will bring in roundabout 80,000 tonnes to 100,000 tonnes of more copper during 2019. We have our zinc assets, which you all well aware, during the lower zinc price, we put them on care and maintenance and we'll bring them up at the right point in the cycle where we believe it is the right time and it will not cause a negative effect on the zinc price when it comes into production. But there, we have roundabout 500,000 tonnes of zinc to bring back online. Plus around a lot of our assets, we have Brownfield expansions which we could bring on cheaply without big capital expenditure and without getting a surprise on cost and time factor as you do get towards greenfield-type operations. So we have a lot of significant growth prospects around our existing assets without having to enter into building a new greenfield mine. We spoke about it that Glencore has spent a lot of money over the last few years on expansionary -- expanding its mines and rebuilding its asset base. So we have a well-capitalized asset base, where we spent roundabout $38 billion expansionary mines since 2009. Today, you'll see our CapEx, including expansionary CapEx and sustaining CapEx, is roundabout $4 billion per annum. And we believe there's no reason that should exceed those levels. Steve spoke about -- if you look at maximizing value creation through capital allocation, we went through that scenario, what we will do with the cash flows, what amounts will be put as a minimum dividend. Unfortunately, Steve has limited me on my $20 billion div by capping the net debt at $15 billion to $16 billion, even though he is happy with two times net debt to EBITDA. So that cap's been put in place there. But that's definitely putting the company in a more conservative strong balance sheet going forward. And that's the way we intend running this company, with a far stronger balance sheet and leaving it at a maximum two times net debt to EBITDA with maximum net debt of $15 billion to $16 billion. Once again, the resilience of the Marketing is proven, and as that slide tells a thousand words, you can see the resilience of the Marketing no matter when you saw the commodity prices drop. We're still pretty well sustained and as -- on the Marketing. And that's why we talk about that range of $2.2 billion to $3.2 billion on the Marketing. Even with lower commodity prices, we will still be within that range. Naturally, we've always said with higher commodity prices, the Marketing will perform better, more arbitrage opportunities, better margin opportunities, and that is well proven by the first half of this year where you see the amount of the profits that we've made in the first half on the Marketing at that $1.4 billion level. Steve spoke well in detail about the -- at current spot prices, the cash flow generation this company generates with current commodities, and we've broken it down per sector. And you can see the generation of cash is extremely substantial, whereby we've got, at current spot prices, EBITDA roundabout $15 billion, generating free cash flow of $7.1 billion, and that is at current spot prices. And if the scenario -- and you have increases in the commodity prices, that will even get higher. So I think the company is well set for the future, generating substantial cash. And we will -- how we allocate that cash will be carefully maintained, and per the slide, we'll set it out where we set out the minimum dividend and where we can see potential M&A or growth opportunities. Otherwise, it will be kicked out in the form of dividends. So I think that gives a scenario of the first half performance and how we see the future of the company going forward. And I think with that, we open to questions.

[Operator Instructions] We will now take our first question from Ian Rossouw from Barclays. Please go ahead, your line is open.

Just two questions for me. Firstly, just on -- I mean, given your net debt is now in a much more comfortable level, it looks like there is a slight change in the tone around growth projects, as you mentioned, material growth projects and options on copper and zinc on the front page. I mean, obviously, outside of Katanga and Mopani and the zinc mines sort of restarting, could you provide a bit more details on what other projects you have in mind and likely time frames and costs? I'm just thinking in the context of your $4 billion CapEx guidance over the next five -- three to five years and how that could change. And then the second question, just on Koniambo, would be grateful if you can provide an update on the asset; seems like there was some news reports of a water leak which caused the plant to shut down recently. You're losing more than $200 million of capitalized OpEx and CapEx. And I understand the outlook for EVs is quite positive. But obviously, that's further out. Are you willing to entertain further cash burn over the next couple of years?

Let's talk about the growth projects. I mean, the growth projects, we mentioned the ones that I said are there. Those are basically the ones we're focusing on. There are other brownfield operations around some of our assets, improve on the copper, where we could do various expansions, brownfields. But that will be for later or not. Let's first Katanga up and running, Mopani expanded and then we can look at potentially some of those other brownfields around those areas. The $4 billion CapEx, I think Steve took you through that. Around about $3.1 billion is sustaining CapEx, around about there. The balance is explanatory CapEx around some of our nickel assets in the Sudbury region over there just to maintain existing tonnes, so it is really a little bit of expansion over there. I don't know if you want to add more on that one, Steve.

I think just on that, Ian, obviously, the profile, if we just let the current expansionary projects work their way through and bring those tonnes, obviously, you have a declining profile on the expansionary of that $1 billion, will gradually sort of tick down to zero over the next sort of three years or so. But we haven't said, obviously, take down your CapEx. So we have said, by all means, sort of use $4 billion next sort of three to five years. It does, from our perspective, just conservatively sort of create some sort of buffer, if you like, from which we can look to bring some other sort of brownfields and various other smaller projects, for which we're not telling you to model necessarily the benefits of that. So I think for now, we're being conservative. We'll have the investor update, which is always sort of end of November. That will be the forum in three or four months when we go through our longer-term and more medium-term planning cycles and just bases our confidence in prices and where we see the outlook. We'd look to provide that guidance back at the end of November about what our production profile would look like sort of '18 and maybe a few years beyond that, where we see some additional projects, some that may reach approval, others that we'll try and give some sense of what likelihood, maybe even some dollars involved, that sort of base of some of these feasibilities, whether it's $300 million for project A could bring X number of tonnes of copper or zinc or coal or some other things. So we'd look to create that sort of forum. For now, just run the $4 billion in sort of perpetuity, and we think that's a conservative profile. But we'll come with sort of a firmer volume guidance, hopefully beyond just the 12 months. So I think the focus has been more shorter term in the last year or two. Hopefully, we can sort of get out of that and give a longer-term profile within four to five months.

In Koniambo, yes, we did have this water leak recently. It took the top plant down for a short period of time, but the plant is operating. The problems we had previously in the furnaces are fully solved. But it is the ramp-up and the hitches you have in the ramp up. But we believe we do -- we have this Line 1 running. Line 1 is operating at capacity. We believe we should get close or very close to budget. We're indicating around about 21,000 tonnes for the year. I think we'll get close to that for Line 1. We are refurbishing Line 2. We will be completing the refurbishment of Line 2 towards the end of this year and starting ramping up Line 2 during 2018. So yes, the plant is performing. It's naturally having its ramp up hitches. We will continue having that. And that's why we've been conservative on the budget setting of Line 1. But once that's running smoothly, and hopefully we have Line 2 up and running, well, eventually, we believe this plant should hit its run rate, so 45,000 to 50,000 tonnes per annum.

And just on the Line 1 sort of indicative costs. I mean, are you happy with the figures you've given historically just on how it's been forming?

I think we're around about there. We think we have -- when I said the whole plant is up and running. We're following those type of cost analysis.

I mean, the key cost efficiencies come from ultimately being a 2-line operation. So you can't judge Line 1 purely in itself. You need the economies of scale ultimately to be around the 50,000. At that point, you've got a sort of a mid-second quarter all type sort of cost structure in the nickel business.

We will now take our next question from Jason Fairclough from Bank of America.

Look, we're getting lots of questions from investors on your exposure to cobalt. Could you discuss the progression in your mined and marketed cobalt over the next couple of years? How do you see the supply evolution in the market? And then how do you think about the risks if the cobalt price goes too high as this market, I think, triples on your math, Ivan?

Yes. My math definitely shows a difficult situation going forward. What do we have today? We have 100,000 tonnes of cobalt production. You will have us with Katanga adding another 20,000 tonnes. You will have -- potentially, you get -- a bit more can come out of Tenke. Understand, I'm not sure the exact numbers, but I think in the DRC they can potentially add another 10,000 tonnes, around about there. You have the other Kolwezi tailings, where over time, you can most probably get -- I don't know the exact numbers, around about 10,000 to 12,000 tonnes potentially there. I'm not sure the exact numbers. So most of the new growth of cobalt will come from these operations in the DRC. Now naturally, if you look at my numbers, where you talked the world will need just an extra 260,000 tonnes of cobalt based on the mathematics by 2030, yes, it could push the cobalt price too high. And I'm not an expert in the Scientifics of battery making, but cobalt is needed, and I don't know if those numbers are correct or even half of those numbers are correct. Yes, we are going to have to find another -- I believe, another way to make batteries because I don't think the world can supply the amount of cobalt needed on those numbers.

Well, they're trying to move the technology to maximize nickel content to move to the sort of 8:1:1 ratios. There is cobalt -- people are dusting off plans in Canada, U.S., Australia. There will be sort of odds and sods all over the place, even outside of the DRC, with price incentive that justifies that. Markets can prove to be quite responsive. And when you have prices that clearly incentivize that, I mean, Jason, I'm sure you cover another 20 or 30 sort of juniors that are sort of looking for cobalt out there. Maybe there is.

I think coming on the chemistry, where they're looking at it, Jason, whereby they can use more nickel and they talk about the chemistries to evolve from the current 5:3:2, 6:2:2 to 6:2:2, 8:1:1, increasing nickel loading further. So yes, I think they will look for other areas to reduce the amount of cobalt because that is the problem. You can't open a pure cobalt mine. It's more of a by-product from the copper in the DRC.

Scientists are looking at even some zinc sort of introduction in there, which will be fine for us if that was an outcome.

Can I just be a little cheeky and add a follow-on question to Ian's question on CapEx? I mean, most of your peers we're seeing guide up CapEx towards one times D&A. You guys have $7 billion a year of D&A, and you're guiding for 4 billion. And just to come back to this question. I mean, to what extent should we think about production fading in some of the businesses? And here, you mentioned Sudbury nickel. I think that's a key area that needs recapitalizing. What about coal? I mean, if you go out five years, does coal need major recapitalization?

No. I mean, obviously, let's see what C&A obviously brings in. Now we've got the 50% of HVO. So that would need to be pro forma-ed somewhere in there and obviously, part of the benefit of that plant in integrating that asset would be also to gain efficiencies around capital efficiencies and look at what the overall production profile looks like in the future. But coal, if you look at that 38 billion, 66 billion in those slides, a lot of that was spent on coal between South Africa, Colombia and Australia. So there's some pretty well-capitalized operations down there. Now current, coal, of that 3 billion sustaining, give or take, is, call it, sort of 750 million to 800 million of that amount. So if you did have a declining profile, let's just say, and we didn't choose otherwise to look to do some Brownfield's and there are some pretty capital efficient operations and options that we do have. If we just let production profile decline, our sustaining CapEx will also consequentially decline as well. So our 4 billion is consistent with -- or at least the 3 billion plus 1 billion is consistent with the volume profile that we have today. So say, Sudbury, that was another one that you mentioned, if that was just to -- over time, without sort of other reinvestment, the sustaining CapEx itself would also decline consequentially. But again, in one or two operations, there will be needs to do some potential Brownfield investments into the group so as to extend some lives or go into different areas or satellite pits or these things. But, Jason, what we said, the 1 billion these are not sort of 6 billion, 7 billion, multibillion projects. These are all much smaller, $500 million and $800 million. They're going to be done over three or four years. So we've said, by all means, use the 4 billion number as providing that buffer, so that's going to allow us to sort of reinvest in the business. And part of frankly, part of the November update that we are looking towards as we go through all of our planning cycle is we do want to look beyond just one or two years and actually come out with a longer-term profile, say this is where we are in each business, this is where mine lives are, this is where some potential CapEx might be required, this is what the profile could look like. And I think you'll find that that's ultimately all going to be, cash flow-wise, to your satisfaction in terms of some of the guidance that we're sort of giving around CapEx. Now if you look at that CapEx against depreciation ratio, for us, you can't really -- a lot of that depreciation has just come from the share-for-share Xstrata acquisition, 2013, brought in the books at $50 billion, whatever the number was at the time, basis. There are huge assets and capitalization that was put in place at the time. It's not reflective of historical actual cash spend of putting money into the ground that can be correlated with a depreciation profile going forward. This was noncash. This was premiums. This was reserve value, this was all sorts of financial sort of accounting-deemed consideration. It was not a cash acquisition, and that's a huge part -- about at least 70%, 80% of that appreciation today would relate to the Xstrata acquisition at the time. That's now being amortized through the books.

If you look [at it], Jason, around our coal assets, and as Steve said, a big part of that $38 billion expansionary expense which Xstrata did at the time, a lot of that was involved around the coal. So these are long-life assets. Also, around a lot of these assets, as Steve mentioned, is easy brownfield expansions. And you got to remember in expanding these coal operations, it's not massive CapEx. So that's what Steve talks, within that $1 billion of the full really expansionary, [indiscernible] part of that is involved around coal and where we can expand the coal. And also that is one of the reasons that, as you know, we were very eager to get the Hunter Valley assets. There's a lot of reserves around the Hunter Valley assets, which we can process a lot of that material through our existing -- that's why we talk about the synergies -- through a lot of our existing washing facilities and the facilities we have around that massive reserve. And as you all know, the Hunter Valley reserve is a 50-, 60-year life reserve, which can be utilized and brought forward if we see a dropping off from our existing reserves, and thereby we get the benefit of that.

And at the appropriate time, we can obviously be more definitive and more sort of detailed on some of those benefits and synergies potentially associated with that acquisition. And Peter -- so we'll put Peter in front of you guys and he can run through all that at some point.

We will now take our next question from Sylvain Brunet from Exane BNP Paribas.

First question perhaps on Marketing, where the numbers were particularly strong in metals. Based on the outlook you have for premium tightness in copper and zinc, are we right to assume that there's no reason at this point why this performance couldn't be repeated in H2? And on the energy side, if you could help us to think about -- I know you don't disclose it, but the coal versus oil trends to help us on the forecast. That's the first question?

Well, the first question here, yes, you are correct Sylvain, if you talk about the marketing. The marketing in the zinc, copper and the metal side had a very good first half. I don't see any reason why it shouldn't perform as well during the second half, as you correctly say. And as we do say, at higher commodity prices, we do perform better, premiums get higher, it gets tighter. The arbitrage opportunities exist. So with the higher prices, and that's what we always talk about, we will be more towards the higher end as we get towards that area. So yes, second half, if we do believe that the copper and zinc prices stay high or even get higher, it should bode well for the Marketing business, and therefore, we should do well there. The other point, I can't remember, you want to talk on the oil side?

The energy.

Yes.

I mean, it's been -- obviously, I mean, net-net, it was ahead of first half 2016. I would think there's no reason why in its totality that the energy business, made up of oil and coal, shouldn't match first half performance. It was reasonable in oil. It was strong in coal. Early days still in the second half. There's no reason why it shouldn't replicate that.

Yes. I think on the coal side, the coal prices, as you're aware, are looking very strong at the moment. We have Newcastle sitting around about $98. You have South Africa up at $85. So it looks pretty good there on the coal side. And once again, arbitrage opportunities, markets are tight, and we even see it today, where you do see the arbitrage opportunities occurring where Newcastle is very tight. You have potential strikes down in Australia. And therefore, you get the arbitrage opportunity where even South African coal starts moving towards the Far East, and that creates good opportunities on the Marketing side. On the oil side, talking to our oil traders, yes, they believe the second half should be similar to the first half, and we will watch what the markets are doing there to see how they perform.

My second question please on how you think about the portfolio and its further optimization. And obviously, terms and conditions are better probably than most of what was expected maybe a year ago. Do you see opportunity to monetize some part of the portfolio? We've seen some price headlines from royalties. And then how should we think about what could become less core for Glencore going forward?

Yes, I think the royalty stream makes sense. It something the guys worked on, and they're getting together with one of the large pension companies to set that up in a streaming type operation and then eventually use that as a vehicle to purchase smaller junior type companies and bring that into a vehicle. So we'll use that as a growth vehicle for smaller type operations, which we -- which are not necessarily to be held in the Glencore portfolio.

Sylvain, sorry, just on some of these newer royalties and streams. They're less so to do with going -- it's not like we did with some of that streaming on our existing assets that we did last time. These are various royalties that we have ourselves accumulated in various business out of time in respect of either properties that we own or even don't own. In some cases, other people own them. And it's a question of putting all these into a new vehicle. We'll seed debt in, others may seed some cash in. We may hold a stake, and that could be a business where we're monetizing existing assets that no one would have any value, frankly, attributed to within our portfolio. And then we can also, similarly with that vehicle, go and have some capital to go and look to deploy in royalties or streams or things in other third parties, where Glencore could secure some offtake and some benefits going. So it should be one of these sort of win-win-win situations potentially going forward.

But when you do talk about us monetizing, Sylvain, when we do that, we do it to utilize it as a growth vehicle. Like we did with our oil storage facilities, we brought in a partner, HNA, the partner's going to be in there, but we're going use that nicer vehicle to grow that part of the business, and therefore, use equity from both sides to grow that business. The same like we did when we sold our zinc business. We put it into Trevali, whereby we hold a 20% stake in Trevali. We do get all offtake from all the operations. And therefore, it's another growth vehicle whereby certain mines, which may be too small for the Glencore-type portfolio, can grow within that portfolio.

And the very last question, if I may. If you could help us think about some of the synergies around the HVO transaction. I mean, should we think beyond logistics? Or how do you anticipate the integration to playout?

Like I said before when we were answering Jason's questions, we are still working on it. Naturally, we got ideas where it can be obtained. We have various washing plants whereby we can process. And if we do increase the production from that larger reserve, we have spare capacity at our washing plants in the area and, therefore, we can process the coal through those washing plants. We have load-out facilities whereby we can utilize some of our load-out facilities with the expansions at that operation. So we believe we can expand the tonnages coming out of those operations using our current infrastructure where we have spare capacity. So as Steve said, we're going to go through those numbers. We'll get a better understanding exactly how we can do it, how quickly we'll push through extra tonnes. And as you are aware, we don't want to be the ones pushing through extra tonnes in the market to create a negative effect on the price, so we'll also have to take that into account. But in November, when we do the full presentation, we'll have more -- a further update on that.

I mean, given you've got various other public companies involved in this stuff, we still need those to bed there. Yancoal's got to bed down its transactions. They got to do their capital raisings. They got to close out. They're a public company. Rio is a public company. So let all these things settle down, complete, and we should be in a position early next year to show what the integrated HVO looks like in our business. Because it will start coming in and change the overall volumes, margin, cost structure of that business, and we'll incorporate, as much as we can, some of the synergies that we expect out of that operation.

We will now take our next question from Sergey Donskoy from Societe Generale.

I have three questions. Two on these selected M&A opportunities that you mentioned during the presentation. First of all, in terms of your commodity positioning, if you could give us some thoughts around it, you have just signed a significant deal with Yancoal. Let's suppose that the C&A deal is closed as its planned. Do you have more appetite for coal assets? Would you go for more or should we expect your next move to be somewhere else? Would you be looking for expansion in zinc or copper or maybe some areas where you're not present yet, or you think you still have room to beef up your presence in the coal, etcetera? Then second question is in terms of balancing your M&A versus dividends, could you remind us what IRRs are you looking for when you are considering potential M&A transactions? And again, in relation to this Yancoal deal, for instance, would you be able to realize this IRR, if we say coal prices go down to the current forward curve, which is something like $70 per tonne? And final question is on your DRC expansion. It's one of your major efforts, you're spending significant CapEx there. How do you see political risks around possible changes in the establishment? Do you think there could be more resource nationalism that would somehow effect the operations there detrimentally? Do you feel comfortable, and if so, why?

Okay, a lot of questions. Okay, let's talk the opportunities on M&A, et cetera, and appetite and where we wish to go, et cetera. What we've always said in this company, it's not a matter of where we're going to say we want to grow in zinc, we want to grow in copper, or we want to grow in nickel. It's more opportunistically. On M&A, we don't go and look for stuff that's available to try and buy something if it's not available. What happens? We sit back. We wait to see what opportunities. We see opportunities, we believe, better than other parties because we deal with all third party suppliers. We're one of the few companies that have a lot of third-party trading whereby we talk to every supplier, we have a relationship with every supplier around the world, so we get to see an opportunity. If someone is trying to sell something or considering selling an asset, we get to know about it hopefully before others, and therefore, we can deal fast and quicker to get that asset. So it's more opportunistically. And the same when you talk about coal, it wasn't as though we really had big excitement to expand in coal, but when Rio Tinto decided to sell their Hunter Valley coal assets, the opportunity arose. Now it's definitely no secret we're always talking to Rio Tinto to try and merge our Hunter Valley operations together with theirs. They didn't wish to do a merger. And they said, "No, we'd rather do an outright sale." So that opportunity came and we aggressively pursued it, as you well know. And it's well documented, the situation which occurred between us and Yancoal competing for those assets. And naturally, because of the large synergies, we really -- those assets made a lot of sense to us and we ended up doing a good deal, we believe, with Yancoal. We will own those assets 49-51 with them. So that was an opportunity that arose. Do we have a bigger appetite for coal? We'll wait and see. If there are more opportunities and someone wishes to sell assets, and we can purchase them at the right price, and we believe at the right IRRs, and an opportunity comes before an opportunity in nickel or zinc or somewhere else, we will look at it. And that's the way we operate in this company.

Well, I mean, currently, we have actually got a couple of smaller noncore assets that we're looking at selling potentially in coal. We've got Tahmoor over in New South Wales. That's currently being sort of potentially shopped around.

Yes, and we always review our portfolio. And if something doesn't make sense to sit in Glencore, we will look to sell that. And as we said, we were looking at Yancoal. We're looking Hunter Valley. We're looking at it to reposition our core portfolio and sell off at least $1.5 billion, we said, of potential coal assets. So we work opportunistically around the situation, and that's how we look at M&A. Then you talk about balancing M&A with dividends. Steve's got a slide there where he shows how we will, what we'll do with the cash flow, how we look at M&A, how we look at dividends with the minimum dividend structure that we set up. But generally, we like to look at the current moment. If we look at M&A and M&A opportunities come around, we would like to get an IRR of around about 15% over there, which leads to your next question. Do we believe that getting coal at Hunter Valley operations, we will be achieving the IRRs we're looking for based on the forward price curve? I don't know how far you have taken the forward price of $70. I think the forward price is higher than $70. It's around about $78 today, the forward curve. And based upon that forward curve, and based upon the synergies we will be getting from this business, we believe we should be getting there. So yes, naturally, if the forward curve drops further, it depends what synergies we get out of this business. But we are being conservative when acquiring these assets based upon our forward prediction and not just looking at the forward curve. So therefore, that's the way we'll look at that. Regarding the DRC and the expansion, where we are going ahead with the expansion, we nearly finished the Whole Ore Leach program at Katanga, the first part of it, and that will bring Katanga in to operation at the end of this year. So the money has -- most of that money has been spent. The balance money on the second line will be spent later on to bring the balance on to spend in '18, to bring the next 150,000 tons in 2019. So yes, that will happen. Now of course, with these expansions, the potential political risks to the DRC, we got to monitor. And as we said, President Kabila has said he will hold elections towards end of the year or early next year, whenever that may take place. Hopefully, there will be a smooth transition in the country. But naturally, we've got to monitor what's happening in the country all the time. However, you have to note, we are in a very southern part of the country. We're in the Kolwezi region, which is far from Kinshasa. And hopefully, if there is any unrest, it won't go towards that region and therefore, it should -- we should have a clear -- we should not have problems in that region. However, you should note, I mean, if there is major problems in the DRC, and we do have problems there, it would have an extremely positive effect on the copper price. You are taking around about -- you take about 1 million tons of copper out of a 22-million-tonne market, can you imagine what that would do to the copper price against the balance of our portfolio? So net-net...

And cobalt.

And cobalt. So net-net, not good for us. We don't wish to lose those tonnes. We don't wish to have any political unrest in the country, and we hope it's stable. But overall, the company is pretty well hedged against something happening there because the benefit we'll get on the balance of our copper will be substantial.

We will now take our next question from Myles Allsop from UBS Bank.

Just may be sort of following up a little bit on those last few questions. Within the DRC, I think the near-term risk is probably around these potential tax increases that we've talked about earlier this year. I guess, Parliament comes back in September. I mean, the industry fought aggressively and won back in 2015. I mean, are you concerned we're going to see sort of a meaningful increase in taxes to fund the election because the government hasn't got any money? And the second, just going back to the M&A versus dividends. I mean, with management owning 30% of the equity, isn't there a temptation to push more money towards dividends? But then, alongside that, I mean, we're -- obviously, Bunge was talked about earlier this year. And could you run through the rationale again for Bunge and why that's such compelling deal?

Okay. Look, tax increases at DRC, let's hopefully hope they don't need them with a higher copper price.

And cobalt.

And cobalt price. And the tax generation to the government should be extremely substantial. So hopefully, they won't require tax increases because the amount of cash flow to them is going to be far greater than before. But naturally, tax increases, you talk DRC, we see it in other countries. It can occur in any country. It could occur in Zambia. It occurs in Australia, where you get various changes to the tax regime. So yes, that's something we've got to watch and monitor.

I think it was more around nonexisting operations and future concessions.

Yes, I was talking about new operations or stuff like that. But there's always rumors of it the DRC. There's rumors of it Zambia. There's rumors of it various countries. But hopefully, I think the increase in commodity price should curtail the pressure on that area. Regarding dividends and management, interesting question. I don't think [indiscernible]...

The management would be very happy with the minimum next year of $2.5 billion.

I haven't seen management push for dividends over and above what we've set out, and I've never seen management push for dividends different to other shareholders, because management would also like to see the growth of the company, which has a positive effect on the share price. Hopefully, management is not short of cash where they really need the dividend short term. So we therefore...

15% IRRs are good.

I think they like the 15% IRRs. And the future would be exact as any of the other shareholders, if we can grow the company better with acquisitions rather than kicking out dividends, which is good for the long-term growth and the long-term share price of the company. And as you will note, most of the management or definitely the senior management have not sold shares. So they're very focused on a higher share price and return on the total equity portfolio rather than pure cash. So I think they will follow the same policies that we have set out, which is for the benefit of all shareholders. Regarding Bunge and the approach, naturally, we can't talk about it, but I can talk about the ag business. We've always said in the ag business, we set up the structure with our partners, the Canadian pension funds, whereby we have this 50-50 arrangement with them in Glencore Agriculture. Now our deal was to grow Glencore Agriculture and have this partner, in the same way as we set up Xstrata in the early years, to set up a vehicle where we would not be doing all the funding to grow in the mining sector. At the time, where we had the public owning 60%, Glencore 40%, we could leverage that vehicle. And if equity was required, we didn't have to put it up all ourselves. So we're just doing the same in the ag sector, where this vehicle set-up is performing nicely. And we have said we want to grow. We would likely grow in the United States plus other areas. And we will continue looking for opportunities. And as I said, if opportunities arise and we think they make economic sense and they give the right IRRs, then we'll continue looking into more growth in that sector.

One thing I don't quite understand, because the grain market is very consolidated already, but is the issue that it's just not behaving in a disciplined way and that's where, if you become one of the big four, you can really start changing the way it operates? Is that the way that we really see value coming kind of medium term out of ag?

I think it's more just our growth. We have a very good ag business. We're very strong in certain countries in the world. We're not well set in the United States. As you know, in the United States, we don't have much infrastructure to be able to be active in the agricultural business. It is an infrastructure game. You have to own infrastructure, so therefore, you can source from the farmer and then get it through the system to get it on board vessels. And we do need infrastructure. We've said in the United States, we're looking at various different opportunities. There is -- it's not fully consolidated in various areas. There are small operators. There are smaller companies operating in that area, and it's not fully consolidated in particular countries. And if we can see an opportunity to consolidate there and get involved, and definitely need infrastructure in the United States, we continue looking in those areas.

Myles, it can also be driven by synergies, efficiency, some of those things. You saw even at -- when we bought Viterra in 2012, there was a lot of, obviously, efficiencies and business improvements that we're able to drive through there.

Thank you. We will now take our next question from Menno Sanderse from Morgan Stanley. Please go ahead, your line is open.

Two questions, if I may. First on aluminum and iron ore, clearly 2 commodities that the group trades but doesn't have much production exposure in. Clearly, iron ore falls in the wrong bucket that you highlighted earlier. But if there is an opportunistic situation, would that hold you back, the fact that it's an early-cycle commodity, or would you still consider it? And aluminum clearly falls in the right bucket or midcycle. Is that an industry that the group sees opportunities in to deploy capital? And secondly, maybe Steven, one for you on Katanga. There was a press release about an accounting review in the last couple of years with respect to capitalizing some inventories. Can you give some color around that situation?

Yes. Look, for the first one, mid-cycle, late cycle, etcetera. It's not as though we don't like early-cycle commodity. I have no problem with iron ore. It's just a matter where you forecast future price. I'm not saying that you don't -- we wouldn't want to own an iron ore asset. However, it's a matter of what opportunity exists. If you do believe in the early cycle, mid-cycle scenario, you would have to put on your price forecasts if you believe that we are -- that is an early-cycle commodity and the price will go down, if that's what we believe. But it all depends demand-supply, and you'd have to see really what is happening with the supply side on that commodity. So naturally, we would have to put a price scenario going forward which we believe, based on early-cycle commodity. And therefore, we'd have to put that price into the model. And if we can buy something based on our price, our forward prediction on the price curve and giving us the right IRR which we set within the company, of course we'd buy an iron ore asset if it became available at the right price that meets that criteria. The same goes for aluminum. If something is at the right price based on our forecast of the aluminum price, we'll look at it. However, you must remember, Glencore does have fairly big exposure to aluminum today, not massive, but we do own our percentage in Rusal, around about 8% to 9%. I can't remember the exact percent of Rusal, which is the world's largest aluminum producer today, producing around about 3.8 million tonnes of aluminum. We then do own round about 50% of Century Aluminum, which is also an aluminum producer, producing around about, I think, 1 million...

800,000 tonnes.

850,000 tonnes of aluminum today, with the expansion capacity at some of the idle plants, which we'll wait and see what they do with that. We have expansion there. And if we did upgrade aluminum, we would look at it. More than likely, it would grow via Century and their growth requirements which they may wish to do forward. So yes, we could get further exposure to aluminum. And we do have exposure to aluminum. Iron ore, you correctly say we don't. And we'll wait and see, if and when opportunities arise.

It's all about the entry point there.

Yes. The entry-point price, IRRs and where we predict the future forward price.

On the Katanga thing. I will just refer you to Page, I mean, 69 of our financials. We actually noted that obviously -- just that obviously Katanga had made that announcement that they're looking into some of the historical sort of accounting to do with inventories, as you said. And we've said that any potential adjustment that may result from this review is not expected to have a material adverse effect on Glencore's consolidated income position and cash flows. So that's all we can really say, obviously, at this stage. It's within Katanga's camp at this stage.

Thank you. We will now take our next question from Tyler Broda from RBC.

We've already covered off all the CapEx, and the HVO, at this point, I think you said all you can say. Just a quick question for Steve. Just on coal, in terms of the royalty and the price-linked mechanism, can you give us a sensitivity to a price or remind us the best way to look at that? And then, are there specific coals where that is more apparent for you guys?

Well, it's most countries, in fact, all our regions, South Africa, Colombia, Australia. And Australia's by state. Queensland and New South Wales all have different royalties, most of which is actually linked to price in terms of percentage thereof or certain indicators. I'm not sure exactly the regime. It depends on the coal. It depends on destinations a little bit as well. It's obviously something that would have added, obviously, $1 or $2 in terms of the overall royalty regime. It's the best I mean, it's obviously a luxury sort of statement to make...

We like having higher royalties.

In terms of its link. There was some increases in royalties a few years ago certainly, at the Australia level. But I think those have been fairly stable. South Africa actually introduced something a couple of years ago. And even Colombia it depends, it's different from Cerrejon, different to Prodeco in terms of its application. Maybe that's something, given the coal prices and some of the volatility, it's something we could potentially provide some sort of guidance or something on that -- at the update we'll do in November, Tyler.

We will now take our next question from Liam Fitzpatrick from Deutsche Bank. Please go ahead. Your line is open.

Just a few follow-up questions. Firstly, just on the net debt side. Is that 16 billion a hard ceiling, or would you be willing to go through that on a short-term basis? And then two more just on agriculture, which you may or may not be able to answer. In terms of ownership, would you be willing to dilute further from where you are at the moment at 50%? And secondly, if there is a potential transaction, will you only do it on consensual or friendly terms?

First question, Steve, you better answer that.

Yes. No, I mean, what we've said, approximately 6 billion. Now approximately 6 billion I mean, it's not going to go to 27 billion or whatever. Now of course, you can allow it sort of temporarily, 16 billion becomes 17 billion. But it also depends on asset, Liam, that's obviously being bought. Because if we're -- I mean, that 16 billion applies clearly within the business today. It applies for the asset base today. Obviously, let's say, you bought Coal & Allied and then you do some -- there's another transaction in a couple of years. Something happens. So you've got to look at where you are across footprint and manage it accordingly. But I would say for now, it's a pretty robust cap that we've been looking to manage the business around. I'm not saying we're going to die in a ditch if it's very short term. It's not going to be to go up to, say 20 billion and then, well, you sort of say we will get there eventually. But if it sort of comes in and you do something and it's 16.4 billion, I don't want people coming back and saying, well, you said 16 billion or whatever. So of course there's going to be a little bit of flexibility around that. It's approximately 16 billion. And you would expect that we would have to demonstrate and augment that if you're looking to tap into that ceiling, it depends what you're acquiring. Or you're getting cash-generative businesses, is there a positive EBITDA? Is it improving the portfolio quality? All those factors that are clearly going to come through, that will be well understood, obviously, at the time. And they'll have to be obviously very short, and it will be small and would be a very clear pathway towards sort of getting sort of back there. So hopefully, that makes sense.

Okay. Then on the ag side, your question, would we be willing to dilute? Yes, we are. For the right reason or the right logic, of course, we're happy to dilute. You've seen our history. I think when we originally started Xstrata, we owned 40%. We diluted slowly down to about 34%. So for the right reasons, the right practicality, there's no reason we have to own 50% of the ag vehicle. So we'd have to see what happens over time. Would we go friendly or hostile? I suppose we cannot really comment on that.

We will now take our next question from Dominic O'Kane from J. P. Morgan. Dominic O'Kane: Just quick question coming back to Bunge, if I can. I just wonder if you could comment on the nature and the location of any synergies that you guys have identified?

Yes, actually, we cannot comment on Bunge.

Sorry, Dom.

Sorry, public company. We cannot comment on public companies.

We will now take our last question from David Prowse from Morgan Stanley.

It's all been answered before.

Good. So I think that's all the questions. Thanks for the time, and thanks for taking the call. Thank you.