必和必拓 (BHP) 2017 Q4 法說會逐字稿

Welcome, everyone, to our results for the 2017 financial year. I'm presenting from Melbourne; and Peter Beaven, our Chief Financial Officer, joined us from London. Now as usual, before I begin, please note the disclaimer and its importance to this presentation.

At BHP, our purpose is to create value for shareholders. This is at the center of everything we do. Over the last 5 years, we have laid strong foundations that will support shareholder value for years to come. We have reshaped the portfolio, created a connected culture, enhanced our capital allocation framework, reinforced the balance sheet and unwound almost a decade of cost inflation. The benefits of these foundations are clear in our 2017 financial year results. With our continuing and relentless focus on cash flow, capital discipline and shareholder value and returns, we will carry this momentum forward into 2018 and beyond. We will further reduce costs and, over the next 2 years, embed additional productivity gains of $2 billion. We will continue to strengthen our balance sheet with a net debt range of $10 billion to $15 billion in the medium term. We will maintain capital discipline with capital and exploration expenditure over the next [few] years of less than $8 billion per annum. And lastly, we have classified our Onshore U.S. as noncore. However, with value as our guide, we will be patient as we pursue options to exit our position.

But first, our results. Over the 2017 financial year, our performance was strong. Through our productivity agenda and further cost reductions, we have offset lower volumes. Combined with better prices, we achieved a substantial increase in underlying EBITDA to over $20 billion at a margin of 55%. And we have improved operational efficiency and capital discipline and so delivered free cash flow of $12.6 billion. That's our second highest on record, and allowed us to reduce net debt by $10 billion. We invested for the future, and the board determined dividends of $4.9 billion. And this includes an amount over our minimum payout ratio of $1.1 billion.

Before I hand over to Peter to go through our financial performance in detail, let me start with safety and Samarco. Tragically, one of our colleagues, Rudy Ortiz Martinez died at Escondida during the year; and just 2 weeks ago, another colleague, Daniel Springer, was fatally injured at Goonyella Riverside. These deaths are heartbreaking reminders that health and safety must come first in everything we do. And we work tirelessly to make our workplaces safer. And to this end, our safety field leadership program is currently being rolled out across the business; and when this is combined with our technology initiatives, this will drive us high levels of safety across all our assets. The health and safety of our workforce is our top priority. We hold that our workplace can be free from fatality and serious injury, and we will not rest until this is a reality.

At Samarco, BHP remains committed to the recovery of the communities and of the ecosystems affected by the down failure. With the support of both BHP and Vale, the Renova Foundation continues to deliver, to deliver the social environmental remediation programs that were outlined in the framework agreement and substantial progress has been made. Samarco, Vale and BHP are in constructive negotiations with the Federal Prosecutors Office, so as to settle the major outstanding civil claims, and of course, restart remains a focus for Samarco and its owners and is extremely important to the community. However, we will be patient as we work to find the right long-term solutions.

With that, I will now hand over to Peter. Welcome, Peter.

Thanks, Andrew. This is a strong set of financial results. Our ongoing focus on productivity and portfolio simplification means that we fully captured the benefit of higher prices. In the 2017 financial year, we generated EBITDA of $20.3 billion, up 64%. We achieved a margin of 55%, our highest since 2008, and our underlying attributable profit was $6.7 billion, more than 5x last year. As previously announced, we recorded 4 exceptional items related to withholding tax and Chilean dividends, the industrial action at Escondida, ongoing funding for Samarco and proceeds from the cancellation of the Caroona coal license. Including these, our attributable profit was $5.9 billion.

The EBITDA waterfall chart, which I present each period, clearly shows the significant contribution that higher realized prices made to our results. Through our actions, we converted all of this upside to our bottom line. This period, we reduced unit costs by a further 4% on a copper equivalent basis, which supported a $1 billion reduction in controllable cash costs. When combined with the ongoing release of latent capacity, we delivered $1.3 billion of productivity gains this year, taking the total of the past 5 years to more than $12 billion.

All major commodities in our portfolio again made a meaningful contribution to these strong results. Our Iron Ore business generated higher margins on record volumes, resulting in EBITDA of $9 billion. EBITDA from petroleum was up 11% to over $4 billion, despite a 13% decline in volumes. And we look forward to reversing the production decline, and encouragingly, petroleum delivered reserve replacement above the 200% over the past 12 months. And this doesn't include our recent drilling successes or the Trion resource in Mexico.

The productivity push in our coal business positioned us to convert higher prices into cash and even including the impact of Cyclone Debbie, coal created EBITDA of $3.8 billion. And lastly in copper, despite the strike at Escondida and extended power outage at Olympic Dam, we delivered over $3 billion in EBITDA.

These strong performances have led to outstanding cash generation. Net operating cash flow over the period was $16.8 billion, up 58%. And as the chart on the right illustrates, free cash flow of $12.6 billion was the second highest ever, well above most years of the mining boom and its high price environment. It's a remarkable result, yet as I'll discuss in a moment, we have many opportunities for further improvement.

But before I do, let me explain how we've applied our capital allocation framework over the past 12 months. We have a very clear framework and it's embedded in every investment choice we make, but any framework is only as good as the outcomes it produces. So here again, we've shown how we've allocated capital in line with our stated priorities. We invested $1.2 billion to maintain safe and reliable operations. We preserved our balance sheet strength, and under the 50% dividend payout ratio, we paid out $2 billion. With the remaining $13 billion, we invested $4 billion in high-returning organic development. We returned an additional $900 million to our shareholders and consistent with our bias to debt reduction, we allocated the majority to further strengthening the balance sheet.

Over the period, we reduced net debt by almost $10 billion to $16.3 billion. Gearing decreased to 20.6%, and following our bond repurchase, our average debt maturity has been extended to over 9 years. We have also launched a further bond repurchase for up to $2.5 billion. We're targeting short-dated euro, sterling and U.S. dollar notes in order to generate higher returns on some of our surplus cash and yet further extend the maturity profile of our debt.

Our balance sheet strategy has not changed. We seek to maintain a strong balance sheet through the cycle. And over the medium term, we expect this to translate to net debt levels of between $10 billion and $15 billion.

We have a broad suite of high-quality projects which will grow shareholder value and provide excellent returns. But our capital allocation framework means that we only invest in the best of these. Combined with our ongoing focus on capital efficiency, we expect annual capital and exploration expenditure to remain below $8 billion in both the 2019 and 2020 financial years. We have learned to thrive on lower levels of CapEx. In 2017 including both organic development and maintenance spend, we invested just over $5 billion. This was marginally below our guidance, primarily due to strike-related interruptions at Escondida.

For the 2018 financial year, we now expect to invest $6.9 billion, slightly above prior guidance due to the rollover of last year's under spend at Escondida and an increase in Onshore U.S. development based on the competitive economics of the wells we are drilling.

The spend will be balanced between maintenance, improvement and investment for the future. Maintenance capital is expected to increase to $2 billion due to a $300 million increase in deferred stripping at Escondida and the smelter rebuild at Olympic Dam. And higher spend on major projects reflects the recent approvals of Mad Dog Phase 2 and the Spence Growth Option. We're also spending around $300 million on Jansen as we look to complete both shafts.

Let me discuss Jansen further. We're always focused on cash flow and capital discipline and we must build both short and long-term shareholder value. So we are very happy that we have multiple value creating options, which span both commodities and timeframes. The Jansen project is one of those options. Demand for potash is growing. It can provide excellent margins for well-placed assets and we have a large resource, which has the potential to provide a low-cost long life expandable mine. While timing is uncertain, we have no doubt that the world will need new potash supply. And when it does, we believe Jansen is best-placed. But Jansen will not proceed unless it passes our strict capital allocation tests. And as we complete the shafts, we continue to thoroughly investigate all opportunities to further improve its economics. We have done this successfully with multiple options over the past 4 years. We could delay sanction to better time first production, bring in a partner to add expertise, secure offtake or share risk, divest to crystallize value and optimize the design to further improve the investment case. So while Jansen is an important option in our portfolio, there's no rush. And as with every other project, Jansen will be assessed in accordance with our capital allocation framework. There are no exceptions. As we continue to work on improving the risk return metrics of the project, we will not be seeking board approval in the 2018 calendar year.

So now to my favorite chart. I think this chart is extremely important. Higher return on capital is a critical metric and is presented here by asset. Over the past 12 months, our return on capital employed improved to 10%. It's a sharp increase from the prior year, but there is still much more to be done. And this is where we're focusing our efforts every day and why we are also confident about the future.

BHP has assets that provide outstanding returns today. But they can get even better. Western Australia in Iron Ore generated over 25% returns last year. It's a great asset, but can improve further. We can reduce unit costs to less than $13 per tonne. At Queensland Coal, lower costs and further release of latent capacity by increasing truck utilization and unlocking the full potential of Caval Ridge will see an improvement in returns. And our conventional petroleum assets continue to deliver strong returns despite low oil prices. Our pipeline of high-return, low-risk Brownfield projects will offset natural field decline and ensure that conventional continues to make a significant contribution. These assets account for half our capital employed. The improvements we expect are embedded in operational plans and our track record in delivering these should give you confidence.

Across our copper assets, Spence is performing well following completion of the recovery optimization project. But Escondida and Olympic Dam are currently generating returns below the company average. At Escondida, after a period of significant investment to address grade decline, the new desalinization plant is now commissioned and the Los Colorados concentrator is ramping up and bringing incremental capacity of 200,000 tonnes. Supported by improved productivity, we can now harvest cash flows from Escondida over the next decade. At spot prices, we would expect Escondida's return on capital to grow from the 6% on this chart to over 14% in 2018.

Olympic Dam is largely a fixed-cost operation, so the secret to improving returns is to increase volumes from the existing footprint. To achieve this, we're today accessing higher grade ore in the Southern Mining Area and progressing options to debottleneck the current operation. Combined, this can increase copper production to 280,000 tonnes with potential upside to 330,000 tonnes, which would drive considerable improvement in returns. But first, we must secure asset integrity and to this end, have a significant maintenance program this year. So together, these plans will substantially lift returns in the near to medium term in copper.

Now, to the right of this chart, there are 3 assets or projects with negative returns, and these represent 1/4 of our capital. I have already talked about the importance of creating high-quality options to grow shareholder value in the future. And Jansen and exploration both sit in this category. At Jansen, as I just mentioned, we are considering multiple ways to maximize the value of the project from improvements in capital efficiency to better timing first production. And through these actions, we will both reduce risk and improve returns. But the project will only proceed if it passes through our capital allocation framework.

In exploration, our recent successes have derisked future wealth and giving us confidence to continue our countercyclical investment. With our exploration program targeting prospects, that would work economically at less than $50 per barrel. We're confident of this investment delivering attractive returns.

Our Onshore U.S. assets require a different solution, and Andrew will elaborate on this shortly. Suffice to say, current returns on this capital are clearly not acceptable.

As you can see, we're making strong progress on returns, but as always, there's room for improvement. Individual asset plans and disciplined capital allocation give us confidence that we can maximize cash flow and significantly increase our return on capital.

Back to you, Andrew.

Thank you, Peter. As you've just heard, Peter, along with the rest of the organization, is incredibly focused on squeezing the most out of each dollar of existing and future capital. And to this end, we've fundamentally changed the way we think about capital allocation and the process that supports this to make sure this discipline remains entrenched throughout the cycle.

So, on this note, let me now turn to our views on markets. Our economic and commodity outlooks remain largely unchanged, certainly as we go further forward. Today, we've released our prospects blog, which is up on our website, and provides detailed insights on how we see the markets. We increasingly aim to communicate with investors via digital and social media. BHP is a company of the future and our economists are ready and waiting to dialogue with you all through that medium.

But let me first make a few quick comments before I discuss our strategy. We continue to believe that population growth, continued urbanization and better living standards will increase demand for our commodities over the long term. And we remain especially positive on the outlook for oil and copper, where field and grade declines will result in a sharp reduction in their base supply. The near term, however, remains uncertain. Some of our commodities are trading above industry marginal costs which cannot continue indefinitely. Nonetheless, we do expect a number of our commodities to perform well over the coming year. And where prices do come down in the medium term, we expect it will settle comfortably above their more recent lows.

We're well placed for these conditions. As I mentioned at the outset, we have laid strong foundations to grow value and support shareholder returns for decades to come. We're simpler, with half the assets we had before and a portfolio now truly focused on Tier 1 assets with common characteristics. We are more connected with an operating model that allows us to fully leverage our expertise and create a workforce that acts as a global community. We now have multifunctional teams that connect across the organization globally to share best practice; more importantly, to make us safer; and above all to solve problems together every day. We are more efficient with unit costs down more than 40%, and that's what kept margins strong. And we're leveraging technology, because the scale of our ore bodies allows us to maximize returns from numerous technology breakthroughs, and we intend to be an industry leader in this area and are investing sensibly to capitalize on this great opportunity.

It is distinctive enablers such as those that allow us to maximize value and returns from our assets, and let me share but one example of a myriad. Our team at Jimblebar in Western Australian Iron Ore identified that Cerro Colorado in Chile had reduced its use of explosives in blasts with a minimal impact on fragmentation, how much rocks get broken. The technique was replicated at Jimblebar and saved almost $4 million per annum. Examples such as this would not be identified and acted upon so quickly without the engaged and connected workforce that we have now created. And as I have said, there are many, many more as well.

And earlier this year in Barcelona, I outlined how BHP would put our strategy into action. And in 2017, we continue to make great progress against the plans I outlined then. Firstly, cost efficiency. Our productivity remains strong and is getting stronger. A further $1.3 billion this year takes our annualized gains over the last 5 years to more than $12 billion.

Secondly, latent capacity. We've made small, high return -- in some cases, very high return, low-risk investments in our existing assets that fully utilize our installed infrastructure and make the most of what we have.

Thirdly, major projects. We completed 2 development projects over the year and approved the execution of 2 more after significant reductions in their capital: Mad Dog Phase 2; and just last, week the Spence Growth Option, both of which represent capital efficient, countercyclical investments in attractive commodities.

Fourth, exploration. We've had positive drilling results in the Caribbean and in the Gulf of Mexico. And additionally, the successful bid for Trion in the Mexican Gulf of Mexico provides us with further opportunity. Just this month at Wildling, we encountered -- that's in the U.S. Gulf of Mexico, we encountered oil in multiple horizons, which appear to be in communication with our neighboring discoveries at Shenzi North and Caicos. And while it remains early days and appraisal is still underway, there are encouraging indicators of this being a significant commercial discovery. Taken together, these successes have the potential to add valuable reserves to support our conventional petroleum business in the years to come.

Next, technology. Our global technology function has rapidly deployed high-value capital efficient programs to unlock resource and lower costs even further. Now, as Peter said, I will turn to -- and finally, Onshore U.S.

This business was free cash flow positive over the year and we are proud of this achievement. But as Peter has mentioned, some of our assets and projects have not delivered the return we or our shareholders expect. Peter has touched on the detailed plans for many of our assets. So let me now take a moment to discuss Onshore U.S.

As I've said previously, the shale acquisitions were poorly timed. We paid too much, and the rapid pace of early development was not optimal. When we entered the industry, our objective was to leverage our systems and scale and become an industry leader in shale, and then replicate the opportunity around the world. However, following a global endowment study about 2 years ago, it became apparent to us that the opportunities to replicate U.S. shale oil elsewhere did not exist. So since then, we have quickly improved our capability within the U.S, significantly lowered investment levels and we've also reduced our footprint through a series of divestments as well as optimizing our remaining position through a number of acreage trades and swaps. This sharpened focus informed our regular portfolio review and we have now concluded that all these shale assets are noncore.

We've used extensive input from independent advisers, and we're now pursuing options, several of which are outlined in this slide, to exit our quality acreage. However, all options will take time. We will be disciplined and use this time productively to maximize the value of this acreage through larger completions, acreage consolidation and midstream solution in the Permian, gas hedging in the Haynesville and further well tests to assess prospective resource across all the fields. We will be guided by value. We know and will know what the acreage is worth in our hands and are prepared to be patient.

So to conclude, our investment case is built on cash generation, capital discipline and, above all, value and returns. In the 2017 financial year, we made great progress. We delivered. We delivered free cash flow of $12.6 billion, our second highest on record. We delivered a $10 billion reduction in net debt. We delivered cash returns to shareholders of $4.4 billion and we delivered an increased return on capital employed to 10%.

While we're heading in the right direction, our plans for the future will deliver much more than this. And in 2018, we expect further strong free cash flow, average returns from our development spend of 20% and, at 2017 financial year prices, another step up in our return on capital.

Beyond this, we'll continue to maximize cost cash flow, maximize cash flow with further cost reduction. We'll always maintain discipline with a focus on only highly capital-efficient investments and, I'll repeat, value on returns will remain at the heart of what we do every day.

We have delivered a strong result today, which reflects the consistent and disciplined execution of our strategy, and we have everything in place to build significant value well into the future.

Thank you. Okay, I think we'll wait a little bit and I presume the lines are now open for questions.

Your first question comes from Paul Young from Deutsche Bank.

First question is on the CapEx ceiling. It's great to see the CapEx ceiling of $8 billion for FY '18, FY '20. And Peter, glad to see the approval of (inaudible) at last. Question, does this CapEx ceiling account for only the high-returning plus-15% IRR projects, so i.e. does this exclude spend on Jansen over that timeframe? And then also, I have a question on just the cost out. Can you break down the $2 billion of productivity gains by division? And just further to that, Andrew, clearly, there is a huge opportunity in frac automation and production creep in the Pilbara beyond Jimblebar and also in the Bowen basin in particular. Can you provide some guidance on just your vision for automation, frac automation in both basins and also about the opportunity to creep production, particularly in the Bowen Basin?

Okay. Look, the $8 billion ceiling includes all capital and exploration, Paul. It's not just selected projects. I don't know, Peter, if you want to add anything to that part and then I'll come back on the cost out.

Yes, just a question on Jansen, it is included, Paul, but as I say, it's just the completion of the shafts that we are assuming at this point in time. Obviously, if -- Jansen won't go forward unless it passes the capital allocation framework. And so, nothing's changed there.

Paul, you're right. I mean, Peter mentioned that we see potential to take our costs in the Pilbara to significantly below $13 per tonne. And next year -- or this year, we expect to do better than $14. And there are equivalent benefits that we think will come through over in Queensland as well. Automation does play a role, and we are pushing that hard, not just of our trucks but also of our drills. And we see a steady ramp-up in the use of automation over the next 5 years, which will enable further cost reductions. But the achievements that we expect -- have made and we expect to make going forward, are much more broad than that. We have set ourselves some very high targets for the availability and utilization of plant and equipment, truck hours if you like, high targets on the cost of maintaining plant and equipment. We have a big drive going on through -- the way in which we handle maintenance, and we also have some new activity as well and making the efficiency of all capital, including some of the minor capital programs, more efficient.

But we continue to work on our culture; and therefore, the approach of our people to the work and their own personal productivity. We work on the efficiency of our supply chain with our new global supply organization. And we have always had a leading marketing organization. But we believe there are more efficiencies that are present -- possible there. So we've always worked in the areas of equipment productivity, people productivity, supply chain productivity, capital productivity and marketing productivity. And we have renewed vigor in this area. And a lot of that is enabled by the fact that we now have a highly effective enterprise-wide system that we use for almost all functionality of our business. But particularly powerfully in the area of work management. And I would have to add that of course, above all of that, the most important target we have beyond productivity -- and before productivity is to make further improvements in safety and eliminate forever the need for me to start these presentations with the communication of a fatality or a life-changing and serious injury.

Andrew, I appreciate all that. Just as far as all those things you're mapping out there, obviously, related to production creep in the Bowen Basin, in particular. Beyond Caval Ridge, I mean, you outlined quite a few very high-returning projects in May at Blackwater and Daunia, in particular. Where do you think this will relate as far as opportunity to creep production and to what level in the Bowen Basin?

Well, we've given some general production guidance for this year, which says that we will increase, on a copper equivalent basis, 7% year-on-year. Now we had a difficult year just past particularly because of the strike at Escondida and some of the outage at Olympic Dam, although we are having a major shutdown at Olympic Dam this year. I would rather get back to you or have Adrian get back to you on some of the more details, as much as we're able to provide on the Bowen Basin itself.

Your next question comes from Menno Sanderse from Morgan Stanley.

Two questions. First one on Olympic Dam. Obviously, the opportunity is big, but the company had to deal with some issues of your own making and some issues that you couldn't control. Can you just help us understand how this investment you're going to make is generally going to stabilize this business?

And the second point comes back on the $8 billion ceiling. It is quite a significant step-up over $1 billion versus what we do -- what's coming especially due this year. What type of projects are included in that? Because for the Iron Ore, replacement mines are kept in the maintenance CapEx. So I'm failing to see what's going to bridge that $1 billion gap?

Okay. I will answer the second question briefly and then, as with Paul's question, hand over to Peter for some of the detail. Well, all projects are in the $8 billion, including, as Peter mentioned in his presentation, quite a step-up in our investment in shale, which we can explain why. So -- but I'll let Peter go through all the details as the Chief Financial Officer. Sorry, I've already forgotten. What was the first part of your question again?

Olympic Dam and how this is...

Yes, Olympic Dam. Well, yes, we really are determined to put the asset integrity, and therefore the operational efficiency, of Olympic Dam on a different path. Jacqui McGill and her team have made good progress but still not sufficient. And we are taking a very large, I think it's almost a 5-month shut, to correct a number of problems and really get that set for the long term, which will then justify the investment that Peter described that we go into the very high-grade Southern Mining Area. We further improve the efficiency of the bottleneck, which in the Olympic Dam is a hoist, so that we can get our production up from, on a good day, 200,000 tonnes a year to nearly 330,000 tonnes a year. At relatively little increase in costs; and as Peter said, most of the costs are fixed. We think this is still a great opportunity for the short, medium and long term to grow copper production into a copper market that we think will see quite a shortage of new greenfield projects coming on in the early part of the 2020s and is targeted to do that. But as always, Peter may want to add to that as well, but maybe Peter, if you start off with just a little bit more detailed breakdown of what sits under the $8 billion. I think you'll probably say this, Peter, but I just want to add, that is not a spend up to number, Menno. We will always hope through our push on capital efficiency to spend less than that. It's just giving the market medium-term guidance of a limit above which we will not go. Peter?

So Menno, just in terms of the increase on this year's capital guidance, just recall that, in fact, Mad Dog 2 will be in full go. SGO will be in full flight as we get forward into '19 and into '20. And South Flank in fact, although it's included in the $4 per tonne, it's still a project, and as we said earlier, the $4 per tonne is not smooth, it's in fact a lot less than $2 a tonne at the moment. But that'll increase as South Flank comes on for the next 3 years. We'll start to spend a bit on it this year. I think the other thing is BFX, as Andrew was talking about a moment ago, this is the expansion of the big debottlenecking of Olympic Dam. That is a bigger project than some of the other debottlenecking, typically, our debottleneckings are about the 200 mark. BFX is going to be bigger than that. There's a fair amount of work that we need to do. It's a great project, no doubt, I think. But nevertheless, it's a little bit more than those others. We are going -- assuming that we are going to be exiting shale as Andrew mentioned earlier, but in the meantime, we've got to look after that business and we've got to run it for maximum value. So we should continue to spend I think, wisely, judiciously, carefully but I think we should continue to spend and we are assuming that we'll continue to spend certainly this year, and to extent we still have that business next year, we've made some assumption that we will continue to run it well.

And then, I think the other -- in -- one, maybe just to highlight is that we're going to -- we've still got to replace (inaudible) as a water source in Escondida. We've replaced La Punta Negra and so we just need to put that extra additional modules in the next 3 years of desal water into Olympic -- into Escondida. It's not very much money, but it's part of the increase. As Andrew said a moment ago, it's not that we need to spend up to this, but it's just, that's the -- we're just giving clarity, some guidance, if you like, for the medium term. Hopefully, that's helpful.

Menno, I hope that answers your question. Just a couple of more points from me. Our investments in shale, of course, they passed the capital allocation framework test with flying colors, but they are obviously designed to add value, profitability and marketability to our shale assets. And just on South Flank, we expect a cost per annual tonne of about $30 to $40. Is there anything more we can help you with?

Your next question comes from Lyndon Fagan from JP Morgan.

Great to see the company addressing a couple of overhangs on the stock bang, U.S. Onshore and Jansen. But I'd like to explore those two in a bit more detail. I guess if we start with Jansen, the wording in the presentation says that the project will only proceed if it passes strict capital allocation tests. But I guess with 70% through a $2.6 billion project, it's -- that investment is bigger than Spence. And I guess, I'm still a little unclear as to why we're still going ahead with that. And I guess what changes have been made to limit these type of, I guess, early phase spends? And I guess the next one is just on U.S. Onshore. If we look at the 3 scenarios that the business was being run on, just wondering if they're still applicable or whether you're running as a cash prior to sale? And if it was to be at the merger, whether that would be the U.S. market?

Okay. Let me deal with the first -- the shale one first, Lyndon. Look, our preference would be, I think, to sell the businesses through a relatively small number of trade sales, but there is an execution risk around that. And so in the interest of making sure we can do things relatively quickly, I don't want to eliminate other ways in which we could exit these businesses, including things like de-mergers, IPOs or vending into special vehicles and so on. We'll look at everything in order to decide what is the right way through this. But for now, we think that probably trade sales is where we would prefer to -- what we think we have the best opportunity of restoring and maintaining value for our shareholder. I mean, yes, I can understand why you might say that we should kind of run it for cash, but one of the features of shale, which I think we've grown to like a bit less with time and see it a bit more of a curse, is that the investments that are demanded there are quite procyclical. You have to continue to invest to actually maintain the value of those businesses. In this case, we're very fortunate that we have investments that we can make that we think are -- performed very well on the capital allocation framework. And as I said, won't just add profitability and value, they will add marketability to these assets. But as always, we will -- we review this very, very regularly through the lens of the capitalized allocation framework. And because of the short-term nature of shale, we do that with a much greater frequency right to the very top of the company.

Peter -- I'd like to kind of give most of potash over to Peter. He handled in his presentation. It is a capital allocation issue, but I would just point out that once we have completed the shafts, we will have totally derisked the project. We will have dealt with all the difficult parts of it, and we will only be 3 years away from first potash when we think it's appropriate to make the right kind of cyclical investment. And that's the big question, which will be determined by what -- how it looks. And we've said today that we wait at least another couple of years. And if we need to wait longer, we will wait longer. And over to Peter now, there's lots of things that we can do in that interim to add significant value to what we've already done in potash. But Peter, over to you.

Thanks, Andrew. Look, just couple of things to add. The shafts are underway and in the event that we didn't finish those shafts, what will happen is that, in fact, that they would collapse on themselves and then we would lose the shafts, which doesn't make any sense at all. So we should spend the remaining $500 million or so to finish those shafts probably by the end of 2019, and at that point, we have an option, which is good to go, time well for the market, as Andrew said a moment ago. We have confidence that the market will balance and there will be additional potash required. And so that's really the basis of that money that we are spending. We are allocating today and in the next couple of years.

In terms of the question about what we would change, I mean, just -- I wouldn't comment. So let me just say, what we would like to have is options. It's good to have options. The key, though, obviously in a way we would think is that we would like to have options that are low cost to hold and that have obviously lots of upside potential and so on. So that will really just give you an insight. It's not obvious, but that's certainly the way we would think about holding and adding options to this organization today.

And Lyndon, let me just stress, it's an option, and if it takes longer, it takes longer. And there are many ways as we wait or we think about the option and we gave a lot of examples on the slide as to how we've done things in the past that we can add value while we wait. I would kind of remind you that both the Spence Growth Option and Mad Dog 2, which were sanctioned this year, we were first ready to consider those projects probably 5 years ago, and we've waited and improved and waited and improved. And over the same period, you've seen us exit projects like Browse -- sorry, exit like Browse and Yeelirrie for lot more money than we spent in setting them up and bring in partners. And I could go further back and show on how we've done them. In some cases, of course, you've seen on the slide a complete regroup to some of the things we've just been talking about, how we can replace capital with productivity and things like Iron Ore. Now that's not as relevant here, but what I want to say, we will wait as long as it takes and we will look at every way in which we can to maximize the value of this option. There isn't a commitment to proceed beyond the shafts and there won't be until it passes our strict capital allocation tests. We've been very disciplined as a management team. That's why we're talking today about thriving on capital under $8 billion. It wasn't so long ago that the thriving capital level of this company was much higher and we've brought it down through the discipline of Peter and his team and the culture of discipline that comes from the very top around capital. And that applies with even more rigor as it relates to the potash option, which is still a nice thing to have given some of the uncertainty in the longer term. And given that we do think sometime in the 2020s, we are going to see a requirement in that market for some form of new greenfield production in the same way as we think that's true for copper.

Your next question comes from Clarke Wilkins from Citi.

Just a couple questions. First off, yes, I think it's good that you -- sort of non-core shale asset's being divested. What about the other sort of assets in petroleum that are some non-operated and are also quite immaterial to BHP. Is there any sort of proposal process underway to look at divesting those?

Also, just in regards to Olympic Dam, I think back in May you sort of talked about latent capacity of 330,000 tonnes there potentially being unlocked. Is it possible to just delve in a bit more detail about what needs to be done there in terms of the smelt at the refinery? As to the power constraints we have in South Australia, at least for now at the moment, how can you unlock that sort of latent capacity at Olympic Dam? And what sort of timeframe are you looking at for that?

Okay. I'm going to give Peter the second question on Olympic Dam. He used to run copper and it's a very much a capital thing that he looks after through his chairing of the investment committee. Let me go -- come back to then to conventional petroleum, Clarke. Look, this is a great business, has been a great business for this company for over 60 years. It has phenomenal returns, EBITDA margins of 66%, and we're good at it. We have some of the lowest costs in the industry, about $10 a barrel. And look at the success of this year in terms of exploration. Potentially 2 very significant discoveries, 1 in Trinidad, 1 in the Gulf of Mexico. And getting to the front of the queue as private industry goes back into Mexico to win the Trion bid. And if you put those 3 together, then you have the opportunity to provide competitive options to extend the track record of petroleum within BHP for a few decades to come. And I wouldn't put it more than a few decades, because clearly, one of the questions we ask ourselves is the changing market for oil. A year ago, we wrote a blog where we signaled -- which is now becoming much more widely shared, a more rapid penetration of electric vehicles. And we see that starting to take off perhaps as early as the late 2020s. And starting to have -- add a lot of uncertainty into the market around 2040, 2050 onwards. But until then, of course, there could be a lot of really good made money to be made in conventional oil if you've got areas where you can move fast and the Gulf of Mexico, both sides, Mexican and U.S., and Trinidad have been chosen as fast-moving provinces that when we get something like Trion, like McGovern and like Wildling, we can put them quickly onto production and so on. So this is a business that is a great business and one that we are good at. And we've been pivoting back towards it since we've been reducing our footprint in shale for the better part of 2 or 3 years, and you see some of the fruits of that pivot today. We do...

Quick extension here. I think the question more targeted, I think, the -- that conventional side of business, I think, there is value added in terms of the exploration, et cetera. It's more of it would be the -- your non-operated assets in Australia like Bass Strait, North West Shelf, et cetera, where there's not a lot of growth in mature assets, non-operated. Yet do they really belong in the BHP portfolio rather than sort of your assets in the Gulf of Mexico?

Well, they're very strong cash generators, and we like cash. And -- but what I will say to you is that every year, we do a thorough review of the portfolio. It starts about now and runs the course over the financial year. And we've just kicked this year's review off and we start the review with a fairly detailed analysis of what we're good at, if you like, how we play to win, things like productivity and safety and capital discipline. And some very wide ranging look at future scenarios out to 2040, 2050, how the world might turn out. And then as we get beyond Christmas and through there, we go through the Board cycle. We then -- we direct that at the portfolio. And I can assure you that the assets that you're calling into question that we will examine very hard, as we will many other parts of the portfolio, some of which may be more controversial than others. But everything will get a good going over to make sure they're fit for purpose for the longer term, and we do that every year. And I expect this year's process to be particularly rigorous in that regard.

Your next question comes from Glyn Lawcock from UBS.

Just 2 quick questions. Firstly, just on the balance sheet target, the $10 billion to $15 billion, apologies, one of your peers came out with a target and it caused them all sorts of grief. Everyone expected them to get back into that range. Just wondering, firstly, $10 billion of free cash flow at spot, admittedly spot's high, plus if you sell shale through a trade sale for what the market think it's worth, you could end up net cash within 12 months. Would you sit there or would you want to get back to within the $10 billion to $15 billion? That's the first question.

And then, the second question, while you've announced Spence, it was great, I mean, every metric was worse than what it was around when we were on site 18 months ago, CapEx production, everything. I know that the capital was 12% higher due to FX. Would you consider hedging the currency? Because I mean, this is a 3-year build. If the currency continues to move against you, that return you talk about of 16% could dwindle, and we could end up with another poor investment decision? And also, would you hedge oil given the similar payback nature of U.S. oil versus U.S. gas?

Okay. There's a lot in that. Let me ask Peter to start, because there was a very distinct question on the balance sheet and whether we would run net cash, and may comment a little bit on Spence; and then I'll maybe add something at the end. So Peter, talk about the balance sheet and the $10 billion to $15 billion and the net cash.

So Glyn, obviously the capital allocation framework as always will be the guide as to what happens in a year's time. If we make, as you say, free cash flow at that spot number that Andrew mentioned a moment ago, whatever proceeds we get from whatever assets, in the event that we get -- so we will want to sit within the $10 billion to $15 billion. We wouldn't want to go to net cash and we think that's not an efficient balance sheet. We don't think that's a necessary balance sheet for a company as strong as BHP. So we're very anxious to ensure that we have a safe balance sheet but also an efficient one.

So just on the -- on SGO, a couple of comments on that. I mean, you're -- Glyn, I think the basics of the underlying operating characteristics, the production, the OpEx and so on and so forth, I don't think really anything much has changed. Yes, the FX has changed a little bit. We thought about hedging it and we had a good discussion around that because we want to open ourselves up to all of the conversations we should have on these things. Tends to be, though, that you've got a currency that sort of rolls with copper. I know that these things are slightly -- there's a variation of the timing on these things. But the large -- that we felt there was a natural hedge, but the other thing is that there is a large amount of the costs, in fact, are in U.S. dollars. And so that's the amount that you could hedge and so on practically was a little less. So interesting question, but I think that's a strong project. We don't think it's going to have any issue providing very strong returns to shareholders. It's not by any means a marginal project or anything else like that. Last thing, just on the oil versus gas. Every time we look at a rig, we do it on a quarterly basis, if not more and more common than that. We take a look at the hedge position. We've had a position on gas that we should hedge it and we have done that for the 4 rigs in Haynesville. And oil, it's a little bit more interesting. We're probably trading in the bottom of where we consider the range to be, curves are a little steeper, so you just got to be a little bit more careful about giving away upside in order to protect downside, particularly where it's been a fairly -- the bears have had a good run at that market, so that's kind of how we feel about oil. But every time we look at a rig, for sure, we have a good robust conversation around this.

Thanks, Peter. I don't think I've got anything to add to that. Glyn, so is there any more questions?

Your next question comes from Hayden Bairstow from Macquarie.

Just a question on the ROCE chart. I just want to get an understanding, your future capital is also expected to return to 20%. Should we be thinking about a target of all these assets that you want to get them all to 20%? Or is the cost reduction in [Daunia and Iron Ore able to] lift that bar for your 2022 target of getting the group level to 20% because obviously, you must be getting rid of shale improves you to about 15% straightaway, but then do you need to think about lifting the bar everywhere? Or are you going to accept lower returns for some of these assets that just aren't ever going to get to a 20-plus percent ROCE?

Well, I think Peter can handle that a bit more. But I haven't said very much recently so I'll just make a few comments. We don't have some particular bar, because we strongly believe that there's a lot of variation in risk depending on the projects we're doing. And so we do risk weight everything according to what we think is the likely variability of the outcome and had that discussion just a moment ago on SGO. So that makes coming up with some sort of fixed ROCE very difficult because we deal with a wide range of risks on our projects. Of course, but we do want to push them to high numbers and the capital allocation framework tends to push that. But it is looked at also on a risk-weighted basis. So sentimentally, we're exactly where you are, Hayden. We want to be in that area. But we'd rather not sort of just start get -- going down the line which is of hurdle rates when we think we have such a variety of risk of the projects that we are considering. And we optimize our portfolio in the dimensions of risk-weighted return, risk-weighted NPV and risk-weighted capital efficiency. Peter does that, and both, from time-to-time and then project by project, as it comes through the IC. Peter, you can maybe add the detail.

No, I think pretty decent answer there, Andrew. I mean, I wish we could have everything well over 20% every single year. But we have this thing called price and it's -- it has quite a big impact, so no doubt, as we've also got cost basis, accounting cost basis, which are part of the equation here, so we've got to take that into account. Look, overall, the important thing is, directionally, we know what is really good. We know that that's got to get better. We know that the stuff that's in the middle that really needs to lift, and that's Escondida in particular, Olympic Dam in the nearer term. And the stuff at the bottom, either it's good quality projects or it's shale. And as you say, that's $17 billion out of the 80-odd, we've got to do better on that and so we need to monetize it. It's good quality ground, but it's worth more in somebody else's hands, and we can deploy that capital better in the capital allocation framework, including as Glyn said a moment ago, for sure, returning that cash back to shareholders.

Peter -- on that point, Peter has a good point to add. I've got a passmark from the CFO, just, I think. And we end on the subject matter of returns to shareholders, which seems very appropriate. So thank you for watching and listening, and looking forward to meeting many of you in the coming weeks. Thank you.