Good morning, ladies and gentlemen, and a very warm welcome to Anglo American's 2018 results. I'm Stuart Chambers, the chairman, and it's my pleasure to do so. Just wanted to talk about two things before I hand over to Mark. Firstly, board changes. Philip Hampton retired in December of last year, so he stepped down after nine years of fantastic service. He was our Senior Independent Director, and that role has been passed on to Byron Grote. Secondly, Jack Thompson. He attended his last board earlier this week and will be stepping down before the AGM, and therefore, won't be seeking re-election, also after nine years of fantastic service. Jack is one of our two non-exec directors with extensive mining experience, operationally and on technical side. That's something we're very keen to replace.
We're towards the end of our process to do so, we are expecting to announce something before the AGM in terms of a replacement. Those are the board changes. The only other subject I wanted to touch on, not ununderstandably, is safety. Very encouraging and very reassuring. You'll hear from Mark how we've improved our safety performance again last year to a new record level, 16% better on recordables than the year before, which was itself a record. Very encouraging. As many of you will, if not all of you know, five people lost their lives in our operations last year. That has to stop, obviously. We launched our Elimination of Fatalities task force last year. The design of that, of course, is to penetrate the whole of the organization, every nook and cranny, and discover essentially what underpins these fatal incidents, the fundamental underlying issues.
We've done that. We've got the program. The program is underway, and we're in the middle of executing that. It's key. I think if you add to that the tragedy last month, where one of the industry's tailings dams failed, which you'll all know very much about. That just serves to underpin, yet again, for all of us, safety has to be our number 1 priority. The only goal that makes any sense in this is zero harm. It's a constant vigil. I think with that, let me hand over to Mark to kick off our presentation.
Thanks very much, Stuart. Ladies and gentlemen, we're pleased to be here today, thank you for joining us. Like Stuart, I think it's important to acknowledge the recent Brazilian tragedy. The one thing I would like to say is all of us in the industry are connecting and talking about how we can all play a part in finding solutions for the industry. I'll talk about that a little bit further, but again, I thought it was appropriate just to acknowledge our colleagues and their losses, and on a broader basis, re-emphasize the need to continue to take steps to make sure that we can create a zero harm workplace. That includes zero harm to both employees and local communities. I think that's our most pressing and important priority as an industry.
In terms of our results for the year, the key theme that we'll be talking to is unlocking our full potential. Consistent with our theme, you'll see that we have a photo of a copper mine with some interesting diagrams on it. That was Tony O'Neill's attempt to show that we've become a modern miner. It is actually a photo of El Soldado. The reason that's important is El Soldado will be the first site where we're commissioning our new bulk ore sorters. By the end of the year, we'll be commissioning our coarse particle flotation units as well. For us, whilst that photo may not look that interesting for a mining engineer knowing the background to it's an extremely attractive photo.
In terms of results, I think the important point to make is we're continuing to deliver on what we said we'd do. While 2018 wasn't without its challenges, obviously the Minas-Rio pipeline being a key issue, we've been able and we've seen step-ups from a number of key areas in the business to try and cover that event across the business. The numbers are respectable, I think, in the context of what we had to deal with during the course of the year. The volumes were up, most notably from copper. Duncan and the team really did do a great job. The feedback that we had from the analyst visit in the last quarter of the year, I think reflected on how much progress I think both the copper team have done and how I think we've done as a broader organization.
EBIT margins continued to improve to 42%, up from 40% last year. That reflects both ongoing focus on productivity improvements, I'll talk to that a little bit later, and cost controls across the business and the marketing contribution. Peter and the guys have done a good job in continuing to improve our realized prices against the value in use metric that we use, and again, making a good contribution. Free cash flow was strong. Return on capital employed was a healthy 19%. I think that we're continuing to build a solid foundation for future performance. I've just pressed the wrong button. We're almost panic stations. In pulling apart the performance, we always talk to safety, health, and environment. As Stuart reflected in terms of safety, five fatal incidents.
For us, it doesn't matter how good the results are, no year is a good year if you have a fatal incident or if you've lost a colleague. For us, we do take the time to reflect, try and understand, and the elimination of fatalities task team work is, along with our tailings work, they are the two most pressing projects or important projects for us to continue to work on. For us, and every one of us in the organization, we're committed to delivering a zero harm workplace. Yes, we've made good progress. Again, best performance in 100 years, 16% lower injury frequency rate. At the same time, again, nothing to celebrate when you lose a colleague. In terms of the health front, again, for us, it's an important issue for our industry, making sure that we're creating workplaces that do no harm.
We've got some work to do on musculoskeletal injuries. That is people doing work where there's a degenerative condition. We've got to do some redesign in those workplaces, and that's an important project for us. Finally, on the environmental side, again, the Brazilian incident cast a shadow over the industry. In our case, we had two pipeline leaks at Minas-Rio that we reported in the first quarter. I've already talked about those at the half year. I guess the positive side there is we commissioned or recommissioned Minas-Rio in December, as we forecast at the time. In the first two months, the operation has gone very well. We're up to capacity. The forecast that we talked about for the full year production appear to be on track.
I'll talk a little bit more about the Minas-Rio tailings dam because we've had a few inbound questions. If I could make a simple observation. In engineering terms, we have what we call an upstream lift, where the dam walls are built tailings on tailings. You have a downstream lift where you've got a more purpose-built containment structure. You've got a water-containing facility, which is another level of engineering above downstream construction. Minas-Rio is a water-containing engineered facility. We think whilst there will be debates about tailings dams in Brazil, the Minas-Rio construction is two levels above in terms of its engineering approach. Happy to answer more questions. I think we're in the right place in terms of those conversations and in terms of what we've built at the site.
To be more specific about our current processes, I want to reemphasize, there's been a lot of discussion around what will we do and how will we engage in the conversations around tailings dams for the industry. We will participate and collaborate with all of our colleagues because this is an industry issue. When I talk about Anglo American, if I can describe it in the most succinct way possible, we have our staff on site that monitor and are responsible for the placement of tailings and the management of water on a daily basis. We have, at each site, an engineer or a competent person that has a background in either geomechanics, geosoil mechanics or geotechnical background that is classed our competent person that is responsible for the engineering of and the correct management of the tailings facility on a daily, weekly, monthly basis.
We have an engineer of record that was with us at the start that helped us design the facility, design the process to manage the facility. We have them involved in monitoring our practices and processes during the course of deposition and at least on a quarterly basis, giving us feedback in terms of what they're seeing, what we're doing, and how we're operating. Because they've become part and parcel of the design and the process, we don't count them as independent in terms of managing the facility. We have a central role reporting to Tony O'Neill outside the line that does regular checks on all of our facilities with his team, making sure that everything we should be doing is in place.
On top of that, we have in each commodity grouping, a tailings risk committee that looks at all the tailings facilities and ensures that the correct processes are being applied. On top of that, we have an independent review that's done on an annual basis by another consultant outside the group. From our point of view, those standards were developed in 2014. The development of the ICMM standards came in 2016, and they drew from our standards part of their design. What we're trying to work out now, how we can work with the industry on a collaborative basis and where there may be organizations may not be able to have that level of internal discipline and process, how do we as an industry create some sort of process or resource that can help us make sure that all tailings in the industry are safe.
It's not simply a single company issue, it's an industry problem. From our point of view, how do we play our part in finding an industry solution? I think that's very important, and that's the important message that we want to leave you with today. In terms of the operations, we continue to make progress, in terms of performance. Over the course of the last 12 months, our productivity has continued to improve. Going back to our 2012 benchmark, we've almost doubled productivity across the group, as you can see with the 98% number. Overall production is 10% above where we were in 2012, with around 50% less assets. Each asset that we retained and we've been working with is now on average doing 30% more or is delivering 30% more production than it was five years ago.
In the main, that's been done through operating improvements, process improvements, incremental debottlenecking, and generally running the asset better than we were five years ago. As a consequence, our real costs are down 43%, or as you can see there, 26% in nominal terms. That's been a key driver in terms of our business improvement and margin growth over the last five years, where we've gone from around a 30% industrial margin, EBITDA margin, to 42%. That's despite a 10% drop in our basket price received. From our point of view, we believe the progress has been solid, but there's still a lot more we can achieve because we are not universally or across the board best in class. We're certainly competitive today compared to where we were, but we're not yet best in class, and so there's a lot more opportunity.
If you take the numbers, we've used an independent report at the half year, and we've tried to look through public numbers to see where we rate against our competitors. I might just explain this chart again, because I'd had a few questions where I'm not sure everybody understands where we come from on these numbers. I've got to walk across to the chart. If you look at Anglo American back in 2012, our average cost position, we take quality into account, so we call it leading margin curve improvement. It's actually our break-even cost curve. We're taking the quality of our product into account and coming back with a break-even price, and we do that for our competitors' products as well. It's mainly in met coal and iron ore that you make that adjustment. The other numbers are pretty easy to take into the chart.
We were at about the midpoint on the cost curve, the 49th percentile. In the last five years, we've improved our relative position across all of our commodities, and it's actually revenue-weighted, so we're trying to give you a weighted position. We're at the 37th percentile, so we're mid Q2 on average. That suggests that we've improved our costs quicker than many of our competitors. I think that's the important point. Everybody's getting better. What we've got to do is get better quicker. Yes, we had probably had, in some cases, a lower base to work from, but doing that across the board and doing it in a consistent way is the big challenge. So far, we've made good progress. Now, against our peers, 27%-38%, 36%-45%. Peer three doing better. Some new assets making a real difference.
We've been competitive in or against our major diversified competitors. At the end of the day, there's always a little bit of noise in the numbers. From our perspective, we believe we can continue to improve our position, and that's going to be an important part of the focus in terms of our presentation today. We've made good progress, but there's a lot more we think we can do to continue to improve that position, because I've got no doubt everybody's in the same conversation about how do we get better. In stepping back and looking at the journey we've been on in the last three years, there are three key points we wanted to make. Firstly, in terms of operating leverage, that is how good are you on the cost curve? We've moved to the bottom of the range.
From a competitive position, we think we're well-placed. Doesn't mean you stop. There's a lot more work to be done. In terms of financial leverage, Stephen will unpack the numbers, and we've still got a little more work to do to make sure we've got our balance sheet where we want it. We're certainly in the right area and looking very strong in terms of our balance sheet and what we can do from here in terms of driving the business and doing the things that we need to do to make sure we're delivering value for shareholders. Third is inside the portfolio, and we talk about Quellaveco being a key driver of profitable growth for the business over the next five years. It represents about 50% of that incremental growth.
The thing, Stephen will use the word balance in his conversation and more than simply about the balance sheet. He'll use the word balance in that we've got Quellaveco as a large-scale project that we're executing on. We've also got a number of small-scale, quick-return projects. Moranbah Grosvenor, and a deep bottlenecking. A new ship in Namibia. Chris has got a range of opportunities he can see, particularly around Mogalakwena, and he's doing that work now. Bruce is working on things inside his business as he does the transition through Venetia. We've got a number of small-scale, rapid-payback projects that gives us a nice portfolio of business improvement opportunities. Again, Stephen will unpack that.
For us, it's starting to give us a real balance in the portfolio in terms of what we're doing today, what we can improve, and where we can continue to enhance the portfolio. To put a bit more shape behind that, if I talk to what we've been doing in the last five years. We talk about the operating model. In some ways, it ends up being a grab bag of a whole range of conversations. It is about doing things differently and making sure that the conversation I'm having here is no different to the conversation that frontline supervisors are having with the people that really do deliver the results. It has to be in the context that's meaningful to them. How do we translate that through the organization and make sure that planned work, the objectives we set, get delivered?
That's really the guts of what we're doing and how we're improving the business. We're also now starting to transition into a different type of improvement over the next five years that we think doesn't deliver a lower result. It delivers continuing improvement, but in a different shape. It then starts to focus on P101. When I talked about our operating practices, let's say we were in the bottom half of the performance in terms of equipment. I'll give you an example. Shovel. A P&H 4100 shovel, the best in class was 45 million ton a year. The Anglo American average from its fleet of P&H 4100 shovels was around 22, 23 million ton a year. Our best was around 30. 12 months ago, we set the challenge P101.
How do we have each part of or each major piece of mining equipment or processing plant delivering top quartile performance? We said, "Okay, the benchmark for the shovel, 45 million ton a year." Every one of our operators who was running a P&H 4100 shovel said, "We will achieve that number within three years." In 12 months, Dawson was hitting 50 million ton a year rate of around a 28 million ton starting position. Looking at how we set the places up, how we run those assets, and that type of improvement is being replicated across the portfolio. It looks different in different places, different baselines, different skill sets. That's the approach. We're doing that with all of our critical plant and equipment. P101, 100 being best in class. The objective is, how do we go beyond best in class?
That's where the P101 came from. As Stephen said, a typical engineer's title for a project that should be a lot sexier than P101 indicates. On FutureSmart Mining, it's around new innovation. How do we reconfigure traditional processes to drive step changes in our performance? When I talk sustainability, I'll show you what we think that transformation of our process looks like in the longer term. P101 in the business, we think copper processing, the Venetia acceleration. Bruce is in the room. PGM shovels and productivities, long wall performance are all areas of the business where we see potential for significant change. Moranbah is a good example where we've doubled productivity over the last three years. The question is: How do you go another 30% over the next three years?
In terms of the FutureSmart Mining, bulk sorting, as I said, El Soldado is the first unit. We've got a unit going into Barro Alto and into Mogalakwena. Coarse particle flotation will have its first unit installed this year. Data analytics underlying the way we think about the business and the way we pull apart the data and think how we can improve the business. For us, the journey from 30% to 42% margin was all about fundamental business improvement. The things we could see on the ground that we could get at. Things that we could change. Reconfiguring the pit. Turning section around 90 degrees. All of those types of changes have helped us get to 42%, including the marketing work and the change in the marketing model.
To get from 42% to 45%-50%, P101, the innovation work, and the projects that we're developing across the business, both Quellaveco and the incremental improvements, are all designed to put us in this sort of territory. From where we are today, a 20%-25% growth in the operations and, at the same time, improving the quality of the business expressed through our EBITDA margin, is where we're taking the business. That's on an apples for apples basis. That's what we're all about. That's what we're trying to achieve. With that, I'll pass across to Stephen.
Thanks, Mark.
How much? I'll get this out of your way. There you go.
Mark got to speak about his picture on his slides. I figure it's only fair that I speak about my picture on mine. I just wanted to squash any rumors that's what I eat for breakfast. It's a bowl of palladium grain, and as Chris would tell me, I probably couldn't afford it these days with the price of palladium anyway. Let's turn to the numbers. As Mark said, a really strong set of results and very consistent with what we've presented period on period. That's something that we really are striving to achieve with the business. It's that run rate, the rhythm building on strong fundamentals. Whether that's the operating model or a strong balance sheet, it's that consistent base that gives us that ability to grow and continue the journey in the business.
EBITDA obviously increased year-on-year despite the suspension at Minas-Rio. That flowed into EPS of $2.55 per share and obviously the 40% payout ratio into the dividend and return to shareholders. $3.2 billion of free cash flow and CapEx at $2.8 billion was in line with our guidance. Turning to EBITDA and just the drivers of that, obviously a 4% increase from $8.8 billion-$9.2 billion across the year, again, despite the headwind of Minas-Rio. You can see there the benefit of prices, a currency offset by inflation and Minas-Rio almost offset one another. Turning to the cost and volume, we'd set ourselves a target of $800 million for the year, and we achieved $400 million of that. A little bit disappointed with our progress there. We did in fact achieve improvements of $800 million, but they were offset by headwinds.
In particular, we had above CPI increase in oil price through that March to September timeframe. A few other little costs, particularly in South Africa, slightly above CPI. We had a little bit of inventory build, which should release itself through the first half of 2019. Not making any excuses. It is our job to deliver that target, and we do intend to do so. Turning to the balance sheet, again, the same sort of messages that we've delivered over the last couple of years. We have spoken about this window that we have, this opportunity to fundamentally reset the balance sheet up until the middle of 2019, and that's the point when we start contributing our cash flow to the Quellaveco CapEx in particular. Very pleased with our journey to date. Net debt down to $2.8 billion.
As we've said, we're pleased with the transformation of the balance sheet. That's some $10 billion that we've reduced net debt by over the last four years. Clearly, across all of the metrics, we're in a very strong position, as I say, and really pleased with the progress there. Just a couple of small little footnotes. Obviously, at end of December, we're still sitting on about $500 million of the cash we received from Mitsubishi. Sort of pro forma debt, you could say, is closer to that $3.4 billion. During 2019, with the change in accounting standards, you'll see some of the operating leases come onto the balance sheet, and that's also about $400 million-$500 million. Again, we'll take you on that journey with us as we work through the year.
In terms of returns to shareholders, obviously the 40% payout ratio, again, we've done exactly what we said we would do. 40% payout on underlying earnings resulted in a $0.51 dividend for the second half, and that takes dividends for the full year to a $1 per share. Again, that's just over $2.6 billion that'll take us to in terms of returns to shareholders since we reinstated the dividend about 18 months ago. Obviously, as well as the debt reduction of in excess of $10 billion over the years, we've invested back in the business in addition to the returns for shareholders. Over that same sort of timeframe since 2015, we've invested some $10 billion plus back into the business. Again, that's important in terms of building and sustaining the future as we go forward. How have we done?
This is a slide I like because we all talk a good game in terms of capital allocation, but how are we going in terms of the scorecard? In terms of the scorecard for us for the year, attributable free cash flow prior to discretionary capital of $3.8 billion. Again, quite pleased with that outcome. How did we use that? We reduced debt by $1.7 billion, paid cash dividends in the year of $1.3 billion, and then obviously $0.5 billion in terms of that discretionary allocation. That does include things like exploration, evaluation, CapEx across a number of small projects, et cetera. You will see that increase as we go forward, particularly in terms of Cave Hill, will feed into that growth number as we go forward.
I am going to touch on CapEx now. You will see in the presentation, we've given you some additional guidance in the outer years. I just want to talk that through a little bit so that we are understanding the value in particular that that drives. In terms of sustaining business, sustaining CapEx, we flagged in December that we would see that tick up a little bit in the next couple of years in particular. Part of our priority as this management team is to not only preserve but also enhance the value of the portfolio of operations that we have. That is exactly what we intend to do. Those opportunities present themselves in a number of ways, and that can be cost opportunities, efficiency, productivity. It could be some brownfield expansion sort of programs or some life extensions.
As we realize that higher production rate and generate value, that comes with an increase in the absolute number of sustaining capital. As we grow the base, we grow the portfolio, and we grow production. A good example of that is the one that Mark mentioned earlier in terms of the long walls at Moranbah Grosvenor. Obviously, with the production and productivity improvements that we've had there, that means you work through your mine development areas quicker, you are into the next long wall move quicker, and therefore, in just a snapshot of time, obviously, your absolute number looks a little bit higher, but it is for the right reasons. It is generating increased production, generating increased value, generating increased cash flow, and that's really what we are all about.
As a result of that, and obviously given that we've got things like Cave Hill coming on in 2022, the Moranbah Grosvenor production that I've spoken about, Minas-Rio coming back to full production, we upped our guidance in terms of that sustaining capital for that $2.8-$3.1 in terms of a longer-term number that you should then guide to. For all the right reasons in terms of generating value. In particular, I'll touch on this in a slide or two, the sort of projects and the sort of productivity improvements that we're working on generate really good margins, very attractive margins, very attractive returns.
That adds to that margin journey that Mark spoke about, moving from 42% towards 50%, and obviously in terms of the cash flow that it generates. The big test for us, and this is something I wanted to just spend a bit of time on this morning, is have we spent the money well? Because I know that's sort of a question that came from the December update call that we did. I'm pleased to say yes, we have. We're seeing exactly the same efficiencies that we've seen on the operating side of the business in the capital spend measured as $ of sustaining capital per copper equivalent production. You can pick holes in the calculation if you want to. The level of improvement is so significant that it's hard to ignore.
That's 30% on nominal terms and closer to 40% in terms of real terms. A really good journey. Shouldn't be a surprise because some of our sustaining CapEx and mine development is using the same diggers, the same fleet that we're using from an operating point of view. It's the same sort of influence that's feeding into our capital spend. As I say, as the portfolio increases, we will look to maintain that efficiency. We'll have slightly higher sustaining CapEx in the next year or two, so you may see that just bubble around a little bit, but we're very confident on the journey that we're on from that capital spend and efficiency. Ultimately, measure us on return on capital employed rather than trying to break it down and pull bits of capital out.
That, again, has been a great journey as we've moved from 9%, 11% up to 19%. Sorry, no pun on those words intended. That's been a great journey for us, that return on capital employed. Again, something that we're pleased with the journey and pride ourselves on. Turning to growth CapEx. Again, that will increase a little over the next couple of years, but no surprises. Our spend on Quellaveco in particular is probably the main driver. As you can see on the slide, we do have a number of incredibly attractive brownfield, relatively small scale individually opportunities ahead of us. Again, the main message is here. They'll be assessed and studied with the same sort of discipline that you've seen in the past. They happen over time. Right? Again, relatively small in nature.
They'll come up for approval over time. The capital spend will be over time, and obviously always subject to market conditions and assessment. Just coming back to the cost and volume improvement that Mark touched on earlier in terms of some of the detail and the drivers of that. Obviously, a big focus on P101 and the technology development. We remain absolutely committed to that $3 billion-$4 billion target by the end of 2022. That's the start of 2018-2022. We're $900 million into that if you take last year's actual and then this year's target. I just want to make sure it's clear that that $500 million for 2019 doesn't include the Minas-Rio turnaround. As we've done this year, shall we pulling that out separately? We'll show the turnaround of that separately next year as well.
This $500 million would be on top of that in terms of an EBITDA driver. As we flagged, I think it was at the half year, some of this will be a little bit back-ended in that time frame. Some of the investment we're making in the technology development, the P101, will just take a little bit of time to flow itself into EBITDA and cash flow. That's okay. It's planned. There's programs in place to deliver those initiatives, and Mark touched on some of the things that start to go live through this year, particularly at El Soldado. In terms of this year's number, probably one of the bigger influences is some of that productivity increase that we're expecting to see out of Moranbah Grosvenor from the Queensland Met Coal operations. Just to wrap it up for me, where does this all take us?
I think steady but meaningful production growth over that next four to five years. I think that is something, as Mark mentioned, I love to talk about the word balance at the moment, but it is a balance for us. It's a balance about what we've done with the balance sheet. Continue to focus on that, make sure we're building from strong foundations. The P101 initiatives should drive margins and cash flow again. Again, it enables us to continue to do what we want to do. Again, built on sound operating practice. A lot of you did attend the copper visit back at the end of last year. You can see the benefits of the operating model flow through and a lot of the other sites are going on exactly the same journey. Obviously, built off a strong balance sheet at the end of the day.
Thank you.
Thanks, Tony. Just while I'm setting up my notes, you can look at another one of Tony's very exciting technology slides. Projects. Let me kick away with Minas-Rio. As I mentioned earlier, we have recommissioned Minas-Rio. All the key parameters that we were looking to deliver have been delivered in terms of the startup. We did upgrade the guidance in December on the basis of getting the earlier license for mining than we thought, 6 months ahead of schedule. That allows us to open the pit up we can blend ores, the delivery of an improved product to port is an important part of that process. The unit cost, FOB, is around 21. The range we expect, 28-31. At full rate, we'd expect it to be closer to $25 a ton, maybe with inflation somewhere near 26.
Our break-even cost at full rate still looks like about $35 a ton, is what we're targeting for the product. It's a 67% product, low deleterious elements. If you've got some really tough questions, Ruben's actually in the audience today, we can cover off anything you may want to hear about. On the tailing side. We have an existing tailings dam that's used, and as I described it, basically a water-containing facility. We've got the license to install, we're actually lifting the next level. The last approval required will be then the license to operate when it's completed. We would expect to have that license mid-year, we are making sure that we keep close to the authorities. The federal authorities have already talked to and publicized the new standard. That has no issues for us at all.
We would expect that Minas Gerais will come out with some comments on new standards in the next couple of months. Again, we don't expect that to impact us given the nature of the installation we have. If I could say it this way, if there were to be something that were to impact Minas- Rio, given the nature of the dam, it would literally stop all mining across the country if they came anywhere near what we've built. We don't expect that to be an issue, at the same time, we'll work with the authorities just to make sure that we've got all the key points covered. We're still forecasting mid-year. As I said, the conversation might go for another month or two, beyond that, we think we're in pretty good shape.
On Quellaveco, again, not much additional to report post-December other than good progress continues. The dam, which was built as part of the river diversion, has gone well. Earthworks are going well, when we characterized the project last year, I said that the next 12-18 months was critical. Big earthworks projects and lots of concrete. Once we get through that phase, which I think is the real critical path program, once we get out of the ground and you start seeing metal being erected, I think we'll have navigated the toughest part of the program, we're very pleased with the progress so far. At this stage, the 44-month schedule looks very good. Costs, tight, still in good shape, particularly with the schedule being in good shape as well. We're pretty happy with where we are.
The $1.3 million-$1.5 million estimate for this year for us will be mainly back-end loaded at a 100% basis to $1.3 million-$1.5 million. Our share would be somewhere between $400 million and $600 million in the second half, given we've got the Mitsubishi funds we're drawing down for the first half of the year. I will continue to keep an exploration slide in the pack because as we speak, John and the guys are starting to show some interesting opportunities emerge. In Brazil, I can confirm, or I will confirm that, yes, we intersected copper mineralization. Yes, the grades are at or above, in a number of those intersections, above minable grades that we would think on a larger scale, lower grade type deposit would be minable. The intersections are, in some cases, above 100 meters. I've got to stress that we've got some interesting intersections.
That's all we have at the moment. As a consequence of those intersections, we stopped all drilling after six holes. We secured 37,000 sq km of tenements. We've overflown the tenements, and we've got three or four prospects that we've identified from the aeromag surveys. We're following those up. During the course of the next 12 months, we'll work out what we've actually got. If I could say this, at the moment, it's interesting. It's not anything that we could call a resource. It's an interesting position that will take a little bit of time to unpack. I expect this time next year, we'll give you a much better scope and be able to paint a much better picture of what we have or what we may not have.
I think it's important to let you know that the things we've talked about have been confirmed, and it's interesting. It's not yet exciting. Somebody described me as being excited. I said, "No, I'm interested. I'm not yet excited." I just want to make sure we've got that conversation in balance. We've also been doing some work across the portfolio and in Mogalakwena. The work at Mogalakwena has been interesting. We're looking at extensions to the ore body. The key points to make here is the mineralization appears maybe to be a bit thicker than we thought, the grades may be a little bit better, and the material may be a little shallower. I think Chris is not yet excited, but he's getting close. We're encouraged. A lot of work to be done. Chris and the team are doing feasibility work on potential options.
We are watching the market very carefully as well. Anything we do will be with an eye to the market and an eye on what's right for our shareholders in the long term. We're just measuring our pace and the work we're doing. The results there are also very encouraging. On sustainability, and really, sustainability is part of everything we do. It's part of any conversation we're in, starting with safety right the way through to social performance. It remains a key part of all of our dialogues. If you look at our portfolio from our point of view, we think that the commodities that we have as a consequence of the quality assets. Remember, we're about a quality asset portfolio. What drives us forward and where we focus our efforts is having the best mining assets in the world.
Yes, we take into account the markets that we're selling into and geographies, but it's about quality assets. As a consequence of that focus on quality assets, the major commodity positions that we're in, diamonds, copper, PGMs, bulks, we think plays into the macro trends around growing middle class consumer world. The focus on the environment, long-term, a greener world, and an electrified world. Now, some people thought that I was talking about a Watford supporter when we talked about electrified and exciting. In terms of the big trends.
The FA Cup. Just to keep you honest.
Yeah, I shouldn't have said anything, should I?
No.
It's all about next year.
Yeah.
In terms of the products, in each case, we're building the portfolio. Bruce focusing, getting the business up to 37 million carats with the transition at Venetia. In the copper business, we talk about Collahuasi, we talk about, obviously, Quellaveco, and the longer term opportunities at Los Bronces, and particularly the underground potential is significant. PGMs, as I said, Mogalakwena. I think a really important piece of work for the platinum team is obviously the turnaround at Amandelbult. I was there with Chris for a couple of days, about 3 months ago, and very excited with the work, but still a lot of work to be done. Making good progress. In the bulks, 2 great positions, or 3 great positions in terms of quality and cost competitive position. More work to be done to improve our competitive position.
Again, we've got a good portfolio across the globe. People have asked, so I'll preempt the question on thermal coal. We've actually halved our footprint in thermal coal in the last 3 years. As we've said, we talk to a just transition, so our position hasn't changed. So we're looking at all the moving parts in making sure we're positioning ourselves to have the right type of portfolio that investors want to invest in for the long term. When we look at, if I connect the operating model, the innovation work, the changes we're delivering, we talk to sustainability in the same conversation. As a consequence of the competitive work we're doing, we're looking at delivering a 30% improvement in energy efficiency by 2030. That helps us drive down the cost curve. In terms of reduction of water. Water plays 2 parts.
Yes, it's a cost. Also it's a constraint on the development of our assets, particularly in the long-term desert climates. It provides us a double whammy in terms of improving the performance of the portfolio. That's a critical piece of work for us. Greenhouse gas emissions is about the climate change debate and how we play our part in making sure the world achieves its targets. If I go back to the strategy around trusted corporate leader, I think the conversation for all of us in the industry today is about tailings. I think we've all got to play a part. As we said, whatever we can do to help the industry lift tailing standards, then we'll be involved in all of those conversations.
What do we need to do with our colleagues to improve, I think, is a good example of the sort of approach we believe is right, in terms of the industry. Thriving communities. The people most affected by our operations are our partners in those local communities. I think over a long period of time, we haven't done a good job with local communities, and we certainly changed our approach over the last few years. We know we've got a lot more work to do. For those familiar with the courageous conversations that are occurring in South Africa with the Archbishop, that's the type of approach in terms of engaging with communities that we think is very novel and different. Certainly creating a new dialogue. Healthy environment. How do we make sure that we're doing all things right in terms of the environment?
Are we doing the best we can do with tailings and all those other areas that potentially impact on communities is where our focus is. We've been doing a lot of work. Our tailings work we've done in 2014. It's part of that program. In terms of the business, what we look at and how we think about the future, in terms of margins, driving cash flow. Cash flow, for us as a team is, we think, the effectiveness measure shareholders are looking for in terms of our ability to return today, tomorrow, and through the long term. Margins is a real focus. Getting ourselves to 45%-50% is a critical focus for us because the cash flow then comes as a consequence. Return on Capital Employed measures the efficiency. That is, how well did you generate that cash flow?
It's easy to generate cash flow in this industry over short periods of time. The real measure of your long-term sustainability and the deliverability or the repeatability of that cash flow, look for your Return on Capital Employed. Did you use the capital well? Will the shareholder support you continuing to use capital comes back to this number. We believe that we can deliver better than 20% through the cycle with the quality of the assets and the work we're doing inside the business. Asset life. The one thing I would say that over the next few years, this number will move around a little bit because if we increase our production another 25%, it's never quite linear how that number moves. At 30 years, we're right out there in terms of life of asset.
What I want you to know, what I want you to take away from today is we look at these metrics, and we drive all of these metrics in making sure we've got a sustainable business, that is cash flow, returns, and long-term continuity in terms of the business. In our industry, they're a pretty tough set of metrics to deliver on consistently. Certainly, that's where we're focused. Finally, in terms of the investment proposition, we've been talking to this for a couple, two or three years now, Paul. I notice a few others are starting to talk to the same sorts of slides. Maybe I'll just focus on capabilities today. You know about the asset story, how we've tried to position the portfolio, and what we're doing in terms of return. I think Stephen's done a good job explaining our position there.
For us, the operating model is about creating a culture inside Anglo American that's about being the best. In the end, I'll say today, we're not the best. I can point to many examples where players are doing better than us on an individual basis. As a team, we're determined to continue to take every step to catch and overtake those that we see as the best. In some areas of the business, for example, I think in the longwalls, we are probably right up there. Platinum, I think, again, we're right up there in De Beers, again. There are far more places that we're behind further than we should be, and we're not going to stay there. We're going to keep improving and we're going to keep improving quicker than anyone else.
In terms of FutureSmart Mining, we can't continue to mine as we've mined for 100 years. You can't just keep building bigger trucks. We've got to think laterally. We've got to get out there and think differently. The changes that we're making to the flow sheet is about the changes that will drive the next 100 years of mining. We want to be on the forefront of that because those that lead those changes will be the out-performers in the industry. Third, we make sure that every product we produce, whatever commodity it may be in, and remembering that diamonds aren't a commodity, we sell dreams in the diamond industry. We have to make sure that. Is that right, Bruce? He was getting a bit worried there. Every product we produce, we have to get the best price we can.
That means understanding your customers, making sure you've got a relationship, and making sure that we're creating value for them to ensure that we get the value reflected back in the price that we're paid for our products. With that, more than happy to take questions. Down the front.
Hello. Dominic O'Kane, J.P. Morgan. Three quick questions. The first, it's a bit churlish maybe to criticize a company for doing what they said they were going to, but a 40% payout, given the strength of the balance sheet, does look a little bit skinny versus some of your peers. Can we just maybe map out how we should think about excess capital returns, when, what's appropriate? Is a 40% payout the appropriate payout ratio for this business? Are there any constraints around preferred mechanism for excess capital returns? Is it special dividends, share buybacks? Any constraints around that? Second question, in terms of thinking about long-term capital planning, the industry's had a very poor track record long-term. Well, a very poor track record in stress testing the downside.
With respect to diamonds, could you maybe just give us some thoughts around how you think about the risks posed by synthetics? What's your base case? How do you stress test that into the long-term capital allocation for the diamonds business? Again, maybe some comments on what's going on at Lightbox at the moment. Then final question, obviously, big moves in iron ore prices. There's a sort of structural premium now embedded into high-grade iron ore. How does that affect your long-term planning for Kumba? Potential for long-dated mine life extensions and maybe long-term exploration there.
Given that we didn't expect to get a question on dividends. I'll hand straight across to Stephen to let him do the first one, then I'll pick up the other three.
You sure you don't want to go first, man?
No, that's okay.
Listen, thanks for the question, Dominic. When we set the 40% payout ratio, we did it with a view over time. Again, I'll come back to that balance where we knew we were probably likely to approve Quellaveco. We have a view of what's in front of us in terms of capital spend, balance sheet journey, et cetera. So we factored that into our thinking in terms of that 40% payout ratio. One point I do want to emphasize, we will constantly assess where we are. I mean, that is our job to do, and we will absolutely do that in terms of considering additional returns. We have maybe tracked to where we are perhaps six months ahead of what we thought, given the strength in prices.
If you recall, when I speak about that window we have to reset the balance sheet to the mid-year, I had said, "Well, we'll probably likely have a 3 in front of it. I'd love to think we'd have a 2 in front of it," and we probably got there a little bit quicker than what I thought. I would just come back to the pro forma concept of debt as well that I mentioned. We have got about $500 million of the cash from the sale down to Mitsubishi on the balance sheet. We know that's going to be spent by half year. We know we're going to start our contributions thereafter. The approach that we really do want to take, and we will run a more conservative balance sheet than we have. Clearly, we are sitting very well in terms of all of the balance sheet metrics.
We'll actively consider it, but we probably are almost at the low point today as we look forward. We will actively consider it. We'll consider all forms of returns to shareholders. We'll probably stick with a 40% base dividend. We will stick with that. If we do consider any excess returns, we'll consider the buybacks, the additional dividends, et cetera, as you would expect.
I think the point to reinforce is the careful consideration of these things. We don't want to flip-flop and jump around in terms of the way we go, and I think the issue is getting due attention at the board, as you would expect it to be. It's the right type of conversation. We want to make sure that the calls are right, and it's being appropriately considered. It's being appropriately considered and discussed as you would expect. In terms of capital planning, I'll make one comment, then I'll get Bruce to say a few words. Firstly, in terms of our capital allocation across the portfolio, the diamond assets, and I say assets first, and the business in terms of we think it's a good place to be.
We think medium-term, the market will go into deficit, and we think the way Bruce has positioned the business for success, both short and long term, is right, and therefore, he'll get his allocation of capital on how he shows he can manage our competitive positions. I think we're in good shape. I'll let Bruce speak for both where he is in terms of Venetia and the business and also to pick up Lightbox. Bruce, if you cover both.
Thanks, Mark. Start with Lightbox. Let's just go back a step and remind people why we launched Lightbox. You'll remember we said that the current synthetic offering doesn't differentiate synthetics from the natural product, and that consumers told us that wasn't right, but that they were confused by the offering. We see there's a perfectly legitimate place for a laboratory-grown diamond offering, but we don't think it's the same as the natural business. That's really why we launched Lightbox. We announced Lightbox in June, and we sold our first diamond at the end of September, and we'll only sell 20,000 carats in total by the end of 2019 before our facility in Oregon comes on stream. Since we've started, the Lightbox offering has had a tremendously positive impact on price from a lab-grown diamond point of view.
At the wholesale level, prices have come down of the competitive product between 30%-60%, which is extraordinary given how little we've sold. I think there is already evidence that the positioning of the product as to what it should be, which is not a natural diamond, is already starting to be successful. We'll carry on working on that. I think on the other side of the question in relation to how do we stress test projects going forward, a couple of things. As Mark said, we think supply has probably just about peaked in 2018, and we think that's good, provided we can continue to grow demand. As long as we can keep growing demand for the natural, there should be a positive outcome there.
Secondly, as Mark was at pains, I think, to point out in relation to the asset quality, we have overall the highest quality assets and the highest margin assets in the diamond business. Our ability to withstand price fluctuations is, I think, considerably higher than other people's. I think in the round of those three points is we feel positive about the projects going forward. Of course, we stress test different scenarios on the upside and the downside in relation to all projects, but we are comfortably, we think, within any range of concerns there.
I think also the positioning on Lightbox, I think probably tells you a little bit about our psyche in terms of positioning our businesses. Where we see there's a need to step out and be bold strategically, we'll take those steps. Bruce and the team are continuing to look at the right steps to make sure this is and remains the best diamond business in the world, and that in of itself has to be it. It's a brand, it's a luxury brand, it's a very special case, and we want to make sure we keep it there and take it further. An important part of the business. In terms of structural, the issues with respect to the capital in the longer term, I think, and I tried to say it maybe not as well as I could.
We've got Quellaveco as a material step change in the business. One of the things I think, and I talk about this, is one of the damages or the damage that Minas-Rio did to us as a business was not only the project in and of itself, it also robbed the rest of the business of the incremental capital you need to continue to improve your business. I think the damage was twofold. We're not going to make that mistake again. When I talk about Quellaveco making a 50% contribution to the quality and growth in the business, those other incremental projects, in my view, are just as important. You're spreading the risk, you're actually changing and lowering the risk profile, and you're improving the quality of all parts of the portfolio.
I think taking that balanced approach in terms of capital allocation is critical, I think, again, reinforcing the point through the syndication exercise in Quellaveco, it's a very different company and a very different company approach in terms of capital as well. Jason?
Well, yeah.
Sorry. I don't know it's two. I apologize.
It's okay.
He's a fellow North American, so you can share it up.
Just do it together.
Exactly.
Just looking at it, in a way, the journey is pretty remarkable, Mark. I mean, you've gone from GBP 2-GBP 20 of balance sheet deleveraging. I guess as you start to look to this P-101, is there a risk that the business starts to run out of momentum? I mean, they talk about low-hanging fruit.
Yeah.
Arguably, this was the worst run of the big mining companies when you came on board, and it's not anymore. Does it get a lot harder for the organization from here?
You weren't going to say the best run, but I'll give you that. Jason, I think the way we think about the business is if you look at the operating teams, the business leaders, we look at the things we can improve today, and some of those have got a one, two, three-year lead. Our job, let's say, with Chris and Bruce, is to think five, 10, so that whilst they're delivering the execution on the things we see in front of us, we're thinking about together, thinking about the next step and the next step so that the improvement continues. I think getting the jobs right and getting ourselves thinking about the right time frames and sequencing the business is as critical, if not more critical, than delivering today's improvements.
I think what we're trying to present today is we've done the basics you would have expected us to do. We've thought about the next level improvement at the operating level with P101. We've thought about the next level improvement with the innovation strategies and targets to change our position against our competitors. We're building a portfolio through our capital allocation on even better projects again. Today, it's a 4-tier strategy that's timed over the next 10-20 years that we think will position this company for the long term. I hope that sort of scratches at the query question. It's never smooth. There'll be a couple of bumps.
As I said, this year will be a little bit, not flat, but as a launching pad for 2021, because the innovation stuff that we're putting in place next year will start to deliver in 2021, 2022, 2023. It's never smooth, but making sure you're sequencing the improvements to drive continuous improvement and outrun your competitors is the key.
Tyler Broda from RBC. Just two quick questions from me as well. I just want to follow up with Stephen on Dom's question. Just in terms of Minas-Rio, could you just walk through what happens if there is, for whatever reason, some delay? Say that you don't get that operating license, how long do you operate for? Can you give some color around that? Secondly, just with the news overnight, the coal ban in China. The Dalian ports from Australian coal. What are your marketing people hearing on the ground there about what's happening? Stephen, just to follow up. With the improvement in the company's margins, now being sort of in line or better than peers, does that change at all your view around having a balance sheet with no cash?
Is there a risk that Anglo runs too lazy of a balance sheet? Which is quite an ironic question considering where we've come from.
Let me go. I'll do the first two and give Stephen a minute to tell you what he wants to say. On Minas-Rio. The current dam, in its current configuration, has around 12 months ahead of it. We would expect to see the licensing done by mid-year. That's on the assumption that it'll take the Minas Gerais authorities a couple of months to form a final view in terms of the dam, in terms of their legislative requirements. We don't think the Minas-Rio dam, given its construction, would likely have any material changes given the engineering aspects of the business. We think they will be more focused on other dams and other constructions. Certainly, that's what we've seen at the federal level. We're pretty confident we're in the right place. At the same time, we're in conversation with governments.
We've got at least six to eight months, we think, breathing space in terms of anything we do. The conversations have been positive. We've got 95% local community support. When we delivered the license and the operating license for the mining business, that came six months early. The fact that we had 90%-95% support from the community was a pretty important factor, I think, in making sure we got the support to go forward. I think we've got good support from the local communities. They've been fantastic with us, even when. In fact, our popularity with the community went up after we had the first spill.
What they said is, "Your reaction to the spill was as good as we could have expected." I'm not saying we're happy, but certainly, I think we've done all the right things, and Ruben and the team have done a great job. Ruben, is there anything you'd like to add to what I just said there?
Yeah. Exactly, Mark. One thing that I would like to add is, it's about the process that the public prosecutors put in place to follow up our construction dam or the raising of the first raising. They selected the São Paulo Institute of Technology to follow up the whole process since we granted the LPLI, which is the installation license in January of last year. They have been following the process together with us. We have this comfort and this confidence that they will approve the construction and of course, depending on the legislation, we have probably to add some monitoring system, but it's things that we can do in one month or two. We don't expect any big delay in this sense. Mark is right. I think we have to be close with them to understand what is coming.
In terms of technical solutions or any other big change, it's not there.
Highly unlikely. On the coal ban in China. Look, we've heard press reports. We've spoken to our people in China. They're suggesting that certainly, we're not seeing any impact on our material. There is some question about the validity of what's being reported, but I don't know if it's right or wrong. We remain open just to see what it is. Remember, most of our coal, we don't have a major proportion of our product going into China. It's obviously an important market, but it's not significant. At this stage, it's a little bit scratchy. Certainly, we haven't seen any direct impact at this stage. We'll watch that very closely. That's after talking to people in Shanghai this morning.
As I mentioned earlier, we're probably at the low point in that balance sheet journey and have probably got there a little bit quicker than what we might have thought. If prices stay up where they are, then clearly we'll be generating very healthy cash flows, and that would feed into our active consideration of additional returns. Again, not telling you anything that you wouldn't think is right. If prices continue where they are today, they're probably above our base assumptions. Yes, that would probably generate additional cash flow.
Obviously, feeding off the margins and the work we're doing, that feeds into cash flow, feeds into less volatility, less sensitivity to movements in price, and it does give you confidence then about how you run your balance sheet. Yes, that feeds into that thinking as well. As Mark spoke about the portfolio of it from an asset point of view, we also have been able to think more strategically about our capital structure and again, lift the vision. You'll see there's a slide in there on the debt portfolio. The work we've done there over the last few years also helps de-risk any volatility and movement in prices as well, because we have very little maturing now in the next one, two, and three years. All of those things feed into our thinking. Absolutely, as I say, pleased with where we are.
It will remain under active consideration.
Paul?
Paul, go.
I'm working my way across, Scott. I'm working my way across. Go, Paul.
Just a couple of questions. The first of which is, given what's happened to the sort of PGM basket price, I was just wondering whether or not there is any flexibility in your mine plans in the existing operations to change the focus of which PGMs you're actually sort of targeting platinum, versus for example, the price we're seeing in rhodium at the moment, and whether or not there's any flexibility there. Following on from that, question mark about Mogalakwena. It's not in the CapEx figures that you've put out there on the board. Just again, thinking about what the longer or medium term thinking might be for that operation, given the margins that it's generating at the moment. Then the third question was just around on the diamond space.
Talking about the sort of downdraft that the introduction of Lightbox has had on synthetics, has that had any impact on the price for natural stones? Are you still seeing enough of a differentiation there that is not cannibalizing your existing offering? Thanks very much.
Okay. Let me pick up the two PGM questions. Firstly, I'll ask Chris to make a comment as well, but in terms of rhodium and palladium, obviously, from our point of view, I think Chris, now palladium actually is a bigger revenue generator than platinum for us across the portfolio. I would bet so. Okay. Rhodium's been very good. We have a little bit of flexibility, but basically we're in pretty good shape. What's the proportion now, Chris?
revenue-wise, platinum's 39%, palladium 30%, rhodium has gone from 5% to 13%. A very nice improvement in rhodium. I think Mark is right, is that the assets themselves are quite nicely differentiated. Amandebult is where we get the majority of rhodium from. Clearly, our biggest prill split palladium-wise is Mogalakwena. Whilst we don't have an opportunity to chop and change, actually, as it stands now, we've got quite a nice balance to the metals that are increasing in price quite substantially.
Mog. Mate, keep going with Mog just in terms of what you're doing with the feasibility study or the pre-feasibility.
we keep trying to get Mark to put the Mogalakwena project on his list. We are in the study phase right now. We're doing a study, there's actually nothing we can do now. If we had ZAR 1 billion in the middle of the room, we don't know what to do with it. We are doing a study. My team and Tony's teams are working very closely together. I think we'll have a better understanding by the end of the year. Probably by the end of next year, we'll be in good shape to decide what to do, clearly it's got to compete for capital like any other project.
I think the key thing is we're trying to make sure that we act responsibly in the market. This is an industry that has traditionally overproduced, being balanced, balancing the quality of the project, it's a great project. It's the world's best precious metals mine. Making sure that we're looking at both sides of the equation, I think is critical, I think the guys are going through that process the appropriate way also looking at the resource development options given what we see longer term. I am, or I will knock back the rumor that the reason maybe there were no coal going through China was that they're buying rhodium and palladium and don't have cash to pay for coal. I don't think that's right. We're certainly impressed with the prices.
We're always a bit nervous when it goes too far because what you don't want is switching either. We keep an eye on both sides of the equation. I think that's very important. In terms of diamond price, natural impact, again, Bruce, do you want to just make that point regarding Paul's question?
Paul, we see no evidence at all of any impact of Lightbox or synthetics on natural diamond prices. It's no secret that the sub-$100 goods in the natural world have battled a bit in the last part of last year. That's for a series of reasons that don't relate to anything structural in our view. It's really a combination of oversupply, very significant Forex volatility with the Indian rupee, and all of these goods are cut and polished in India, and difficulties in smaller businesses in the midstream accessing finance. Interestingly, the early evidence from the U.S. over the Christmas season is that the sub-$500 goods did particularly well. We have no reason to believe there's any structural issues going on in the diamond industry in the lower value goods.
Was there any other question, Paul? Okay. Yep.
Ed Miles, also UBS. First of all, could you provide a bit more color on cost inflation in South Africa? Obviously, there's some big power cost numbers being thrown around 16% per year for the next 3 years. Labor costs, obviously platinum negotiations this year. Are we seeing a meaningful pickup now in cost inflation? Maybe for Mark as well, the spot free cash flow after full CapEx, what are we looking at for 2019 at the moment? Then maybe for Chris as well, I was just wondering on the palladium, rhodium, I mean, it feels that price has been pushed by speculative activity. How long do you think before prices crash? Chris, I think the answer is 10 years. Is that Make a comment, Chris, and I'll pick up those.
Clearly rhodium did double last year. There was a very substantial pickup in palladium price. The palladium price is structural, and it's been coming for a number of years now. You'll recall that we've been talking about a deficit in the palladium market for many years now. Clearly everyone, as it was being supplied palladium from above ground stock, everyone said, "Well, we'll believe it when actually the above ground stock runs out." At the end of 2016, for the first time, we really started seeing the structural deficit and the fact that there wasn't material around. You'll recall in 2016 we had that shutdown in our precious metal refinery. We needed to go back into the market to buy metal. Then we started finding it actually was very difficult to buy palladium.
Towards the end of 2016, we started seeing the price move for the first time. 2017 and 2018, over the last, well, the last three years, we've seen 1.5 million ounces of palladium ETFs being redeemed. This is what's been supplying the market. Over the last two years, you haven't been able to get that. We went down another 600,000 ounces in palladium. Last year, the Johnson Matthey numbers looked like they were in deficit. It required the Russians to sell 300,000 ounces from their stock and 600,000 ounces from ETFs just to get to that balance number. The point is actually the structural deficit in palladium with very low above ground stocks is what's driving the price. Is there about to be a crash? We don't think so.
There's also, remember, in a year's time you're starting to see China 6, which is the loading's about a gram per vehicle extra. That's 700,000 to 1 million ounces extra of palladium that's going to be needed. If you look at where that can come from, it can only come from platinum switching. We don't think that there's any likelihood anytime soon that you see a palladium crash. I think there's much more likely that palladium really runs hard, and it drags rhodium with it. Rhodium is still less than twice the palladium price. Rhodium is cheap at these prices. There's been some speculation in rhodium, we think, but actually generally is still an underlying basis for rhodium demand. We know the Chinese have been buying additional rhodium as they've been putting extra rhodium onto their catalysts.
You know, I don't hold your breath for, Mark, 10 years, maybe? Maybe a bit longer. Rhodium, palladium still look good, and there has to be some substitution from platinum back into palladium. We don't think that happens anytime soon. We still think that's 2 years out. The reason for that is because the car makers at the moment are trying to deal with real world driving issues, and that's why they're not substituting yet. I think you can expect to see basket price for PGMs remaining good. Stephen?
I think the key point that we took, we had a very good strategy session and a lot of good questions from the board regarding understand the markets in PGMs. You've got to look at rhodium, palladium, and platinum on a continuum. The basket price is really important to understand because producers of different products, catalysts can move across that continuum at any particular time. What is required to then switch between the two is quite an administrative exercise. I'm talking to people in the last few months about what does that entail. It's not a continuous movement. It's a bit lumpy. The basket price gives you a sense of what funds are being allocated into PGMs in terms of industrial applications.
I think that's the more important number to focus on because those uses are actually, they're quite substitutable, but the administration process to make the change is quite significant. Over the long term, we look at them as a continuum. In the short term, where they go, we provides us with a bit of flexibility in terms of our short-term capital allocation, but we take the long view on the big projects. That's why we think very hard about them, the whole concept in the longer term context as well. We don't know what it'll do in the short term. None of us know how quickly substitution may occur, but we think the underlying fundamentals are very strong, particularly with the potential for the hydrogen economy, the greener economy.
We think we're very well positioned in a business with great assets in what we think is a long-term high potential business.
Mark, I'll just touch on inflation quickly. Just to put it in context for us, the above CPI inflation for the year 2018 was, the impact of that was $200 million. Now, most of that was oil or energy, or energy tariff-related, as that fed through, particularly through that March to September period. They were the main drivers for us in the current year. Mark can speak more knowledgeably than I about Eskom and just where that's heading in a moment. In terms of spot free cash flow, obviously that depends on your price assumptions. What Paul has set out in the pack is a pretty good ready reckoner table in terms of EBITDA, a sort of reconciliation and calculation.
Hopefully you can model Anglo in about five lines, as you go forward just to make your life a little bit simpler in today's world. I would expect You know, as we've discussed, you know, again, depending on your price assumptions, we've set up a nice rhythm in the business, so there's no reason we wouldn't be generating the same sort of free cash flow next year as we did this year. Obviously, if prices are higher, you'd expect some of that to flow through to net cash.
I think the other question there in the middle there was 219 CapEx and deployment. Did I get that right?
Yes.
Has anyone answered that question?
That was me.
You okay? Good.
Yep.
There's another couple of questions.
Two questions.
Thank you very much indeed.
Hi, Sylvain Brunet with Exane BNP Paribas. Two quick questions. First, maybe on the momentum of improvement. Kumba has been on a pretty impressive journey on productivity. What is left in the tank?
Yeah.
My second question is on diamonds. If production globally has generally peaked, how can we explain we're not seeing a better traction in prices? What does it tell us in terms of inventory?
Which-
Inventory.
Diamonds.
Diamonds, diamonds.
Oh, diamonds, sorry. Yeah.
What does it tell us in terms of inventory? Thanks.
Okay. All right. On Kumba, Themba and the guys have taken the break-even delivered cost into China from $77 a ton to around $42, and we got down to a low of $30, depending on lump and quality premium. At these sort of numbers, we're probably in the $35-$40 range, which is impressive. Themba and the guys have been talking about a $10 a ton improvement potential on their base operating costs. He won't get that in a year, over the next 3 years. That would put him ahead of inflation. Some of that will be given back to inflation. On a real basis, that's the sort of number he's pushing at based on the P101 programs and some work on the DMS plant. $10 over 3 years, that's what his objective is.
He'll give probably half of that back to inflation for. I think it's about that range.
Yep.
He gets some incremental improvements through the DMS plant. Beyond that, some of the bigger innovation issues need to kick in more solidly, he's already extended the life to 20 years, we're pretty pleased with the work that's been done. We can still see a lot more to come. On the diamond side, Bruce, do you want to, again, make a comment on pricing and where to from here?
It's a complicated question. I think if you look for a start at the growth in diamond jewelry consumption in the U.S. in the last 4 or 5 years.
Right
You'll find it actually has gone up year-on-year. At a consumer level, there has been increased demand in the main market. Outside of the U.S., it's a little bit more mixed, but it's very much Forex driven, for example, because ultimately diamonds are priced in dollars, and if you live in a non-dollar country, that can be quite difficult for a retailer to match a dollar price increase. If you also look at supply over the last 3 or 4 years, there has been a reasonable increase of supply over the same period. Not necessarily of higher quality goods, but there has been an increase of supply, so there's been more carats available to be sold. We think that peak is about now. It may be in a few months' time, but it's not years out.
Clearly, where there have been difficulties in this period is in the midstream. The midstream has battled, as we've said before, to access enough finance on a sustainable basis to become more efficient, and I think there's more work to be done there. I don't think there's evidence that there hasn't been demand for diamond jewelry over the last 4 years in the main market. We're hopeful that, and we believe very strongly actually, that as long as we can continue to drive demand, which is a big thing, but we're very good at that the tailing off or the leveling of supply for the next few years should be helpful for us.
I think Bruce is probably a bit too polite. Heaven forbid. I think if you look at Argyle, for example, and I'm just using this as for example, and it's not a criticism, it's just a statement of fact. Argyle's in the last 2 or 3 years of its life. It's winding down the clock. The average price per carat, about $28. There's a lot of volume coming out of Argyle at the moment, but that winds off pretty quickly. Certainly not a commentary on what they're doing. I think as they get to end of life, they've got to run it hard to try and hold a margin. That is impacting the small carats, I shouldn't say stone. The small carats. That's having an impact.
That comes out of the system, and you'd be able to model that, I would expect in your models, and I think that's been part of the story as well. Okay. One more.
Two very short questions. Diamonds also. South Africa last year performed not very well in terms of pricing. I think there was about 15% decline in realized price, and this is the area where you are investing heavily. I wonder just what was the reason behind that. Is it some structural change, or should we expect pricing to improve over the medium term? Second question, this slide 30, where you are showing your capability in terms of production by individual commodities. Is this something that we should think about achievable, say, by 2023, which is kind of the-
Run rate type numbers.
Huh?
Run rate. As in run rate type numbers.
Yeah.
Yes, I think that's about the right number, Paul. Yep, 2023.
Okay.
Bruce, yeah, just on the South African.
I'll take the South Africa question. That's not data that we have that supports that question. We certainly had less production in South Africa in the course of 2018, and that was caused to an extent by a stoppage at Venetia following a very unfortunate fatality. The volumes are certainly down in South Africa, but our average index price for rough diamonds for De Beers across the group for the year was up 1%. Venetia's average production is kind of in the middle of the average range. Actually, there was no significant decline in rough prices for Venetia. There was a decline in volume in South Africa.
We can talk later if there's some difference in the numbers. We can help you see where they are. Okay? All right, guys. Thank you very much for the questions. Much appreciated. Thank you for joining us and have a great day.