Vale SA

BOVESPA:VALE3

Earnings Call Transcript 2018-Q3

[Operator]

Good morning, ladies and gentlemen. Welcome to Vale's conference call to discuss the third quarter of 2018 results. [Operator Instructions] As a reminder, this conference is being recorded and the recording will be available on the company's website at vale.com at the Investors link. This conference call and the slide presentation are being transmitted via Internet as well, also through the company's website.

Before proceeding, let me mention that forward-looking statements are being made under the safe harbor of the Securities Litigation Reform Act of 1996. Actual performance could differ materially from that anticipated in any forward-looking comments as a result of macroeconomic conditions, market risks and other factors.

With us today are Mr. Fabio Schvartsman, President and CEO; Mr. Luciano Siani Pires, CFO; Mr. Eduardo Bartolomeo, Executive Director, Base Metals; Mr. Luiz Eduardo Osorio, Executive Director, Sustainability & Institutional Relations; Mr. Alexandre Pereira, Executive Director, Business Support; and Mrs. Marina Quental, Director of People.

I would like to inform you that Mr. Peter Poppinga will not join the conference today. Mr. Luiz Meriz, Global Director of Sales of Iron Ore and Coal, will be available to answer any questions related to the iron ore and coal business.

First, Mr. Fabio Schvartsman will proceed to the presentation, and after that, we will open for questions and answers.

It's now my pleasure to turn the call over to Mr. Fabio Schvartsman. Sir, you may now begin.

[Fabio Schvartsman, executive]

Thank you. Good morning to all. It's a pleasure to be here and to discuss with you the latest results of Vale. And on top of everything, we think that we are moving forward in the right direction, becoming, as expected, a more predictable company showing consistent results during time with a meaningful evolution. And it is important to emphasize that even with the reduction in price of the index of iron ore from the third quarter of last year to the third quarter of this year of 4%, we were able to more than compensate that through our quality -- to our quality premium, actually. This is how the company is dealing with, always, this scenario. And we've been helped by the lower volatility in the iron ore world in comparison to other ores, where our -- where the change in volatility is much more meaningful these days.

I want to make a quick remark on capital allocation. I think that this is the most significant thing for a company in the mining universe. And because of that, we are taking a very cautions look towards anything that relates to capital allocation. Our primary goal is to distribute dividends and to buy back shares in order to compensate our shareholders. And again, marginally, we are making investments as the 2 that we are announced today, the Salobo investment and the Gelado investment, both are value-filled investments that they have a very meaningful return on investment and they basically showcase how we think about investments and how we think that we should handle them from now on.

One quick word in our business strategy moving forward. First, iron ore. Iron ore, we have a very simple equation in front of us. We are helping out S11D. We are helping out the 3 pelletizers that we have -- we started this year. Therefore, our goal is to introduce more high-quality ore in the market, a market that is, today, driven towards higher quality and therefore, the right product for the right market. Meanwhile, it's important for you to understand that we have a clear strategy for Base Metals as well.

Base Metals. We -- as we anticipated since last year, we made, on purpose, a reduction in production. Not reduction in capacity but in production of around 50,000 tons of nickel per year in 2017, 2018 and 2019. Therefore, our play in nickel aims towards 2020 when we will have, at the same time, much higher production. We are aiming to have 310,000 tons of nickel production that is just coming back to our capacity in 2020. We are working towards a total restructuring in our business that will reduce, in a very meaningful way, costs. And this will be present in 2020 as well. And finally, it is effective to see a clear reversion and increase in price of nickel by that time. So while we are going to enjoy the good moment of iron ore, we are preparing ourselves to substitute or to add to it the good performance that Base Metals will show with a jump in results in 2020.

So one quick word about next quarter. Our expectation is to continue to deliver next quarter in a comparable basis. Therefore, with the fourth quarter of last year, we are expecting a very nice growth in results as it happened in the third quarter driven by the recent increase in iron ore prices in the market that is passing quite well. Meanwhile, you have all this noise regarding trade wars, but you see that iron ore is a physical demand and it is working accordingly.

So this is -- this was my introduction. And I will now pass to Luciano to complement with some information.

[Luciano Pires, executive]

Good morning and good afternoon. I will start by -- with costs. You could see iron ore costs reduced from $14.7 to $12.4 per ton so we deliver on our promises. We should expect, for the fourth quarter, this level to remain. And as you are aware, in the first quarter of next year, due to seasonality and the rainy season in Brazil, likely, volumes are going to be lower and costs should slightly creep up.

In terms of Base Metals. I want to point out that not only we have higher unit costs because of lower dilution, because of lower volumes, but also, there was extra spending in the maintenance shutdown for Ontario. So, therefore, more spending over the last volumes and not only last volumes.

In terms of coal costs, although you saw improvement in volumes, the costs did not come down accordingly. And the reason for that is because we are moving a large material in the mine as part of the debottlenecking process. So there's a lot of overburden being moved. This should continue for the next 3 quarters. And we expect a meaningful uptick in product volumes in the second half of 2019. Actually, probably, we're going to be running close to 17 million tons of capacity on the second half. Average for the year -- the total for the year will be lower but the rate of production in the second half will be very meaningful for coal in 2019.

Now moving on to the projects. Some color on Gelado. The investment of $428 million will be roughly half-financed by the savings on replacement of equipment in Carajás because the remaining mining operations of Carajás will be reduced. And in addition to that, because it operates without trucks and 0 transportation distance, the operating costs, the C1 cash costs on Gelado will be $3.50 lower than the operating cost of the Carajás mine, which are already amongst the lowest within Vale. So a very accretive project.

In terms of Salobo III. The expectations to receive the bonus from Wheaton in 2023. The risk associated is very low given that the production targets that we need to achieve. We're very confident it's a sister plant of the 2 that we have already there. Very detailed engineering in place. We see very little risks of achieving the bonus. And even if there is a slippage in terms of ramp-up, the sensitivity of the 5 of the bonus received is very low related to the timing of the ramp-up. We're talking about tens of millions of dollars, so therefore, it's a given that we're going to get a substantial amount of money in 2023.

Capital expenditures also hit a low in the quarter. That's a consequence of finishing some important projects like the emissions reduction project in Canada and the beginning of other important projects such as the VBME and Gelado. And also the comprehensive review during the year, aiming at optimization, capital allocation, also in sustaining investments. So you should expect capital expenditures to go up in the fourth quarter to more normalized levels.

Finally, looking into balance sheet and capital allocation. You saw the 50% sale of Eagle Downs. We have not touched on this before but this is just an example and the proof that we continue to streamline our portfolio, aiming at a more simplified company and bringing in proceeds. We also performed, I would say quite well, in our share buyback. We have almost 50% of it completed in the third quarter alone at an attractive price of $13.27. We were very opportunistic writing the volatility in external and internal markets.

And finally, debt management. We are very close to our net debt target. We reduced overall indebtedness by $800 million. But if you look at debt maturity until 2021, the reduction was even greater, above $1.3 billion. So we continue also to optimize the overall profile of debt.

Having said that, we can jump straight to Q&A.

[Operator]

[Operator Instructions] Our first question comes from Carlos De Alba with Morgan Stanley.

[Carlos de Alba, analyst]

So the first question maybe for Fabio or Luciano. How do you decide between paying dividends or buying back shares? Is there a particular inclination? What should we expect going forward? What are the parameters that we would need to consider in order to understand your rationale? And then second, if I understood correctly, Fabio, you basically suggested that this year, the story has been about iron ore and high-quality premium -- over high premiums for high-quality material. And then 2020, the story would be Base Metal, how are you seeing next year? What is going to drive the company's performance in 2019?

[Fabio Schvartsman, executive]

Carlos, thank you for your questions. First, the discussion between the dividends and the buyback is basically opportunistic. It's a decision that we will make every time that we have a decision on these. And last time, we made a decision to invest in buyback instead of more dividends, first, because we have just created that policy, and we were pleased to pay dividends according to the policy not on a different level in order not to confuse the market. Second, we thought of the opportunity for buying back was a good one. Anything that we -- we're right because we bought back shares around $13 and ahead of the shares today are around $15. Therefore, the return on investment here was quite fast actually. And this is a good thing. That's -- it reinforces that we made the right decision to move in this direction. So it will depend on circumstances, it will depend, say, how is going to be the environment in the fiscal area that impact or not dividends. This decision will be affected by a number of things that will be evaluated at any given market. Secondly, about the 2019. 2019 will be similar to 2018 but better. That means that we will have more of the same in iron ore. We are going to have more high-quality ore coming on stream, more pellets coming on stream. And therefore, our average pellets price should be higher than this year. On top of that, in Base Metals, our cost reduction is in process. So we can -- we are going to see some initial growth results coming out of it. But Base Metals is a story basically for 2020. Where we are really expecting a big jump in performance that we'll achieve the goal that we have for, say, 30% of our result coming from the Base Metals.

[Operator]

The next question comes from Andreas Bokkenheuser with UBS.

[Andreas Bokkenheuser, analyst]

It has been reported that obviously, you're considering investing a bit more in S11D, but obviously, you don't want to bring more iron ore volumes to the market. So effectively displacing some of the volumes in the Southeastern System. Can you talk a little bit more about your plans, if that is in fact what you're considering? What are you thinking in terms of time line? What are you thinking in terms of CapEx? What's required in terms of infrastructure and licenses and so on?

[Fabio Schvartsman, executive]

Andreas, thank you for your question. You're right, we are starting to extract a little more from S11D with marginal investments. We are actually in this phase. So our information is not found at this point and that's the reason why we are not sharing the precise information. But I think that we stand to have a chance to deliver that by Vale Day moment, when we should have terminated all these analogies. And we'll have all the information regarding this. And it will be developed in a very short period of time.

[Andreas Bokkenheuser, analyst]

Understood. If I may then instead answer -- or ask another follow-up question, more on the freight side of things. In the last couple of quarters, you've obviously mentioned that you are in negotiations with some of your ship vessel providers ahead of the whole new IMO regulation that's coming online in 2020, effectively trying to convince them to invest in more environmentally friendly equipment, scrubbers for the ships and so on and so forth. Do you have any update there? How are these negotiations doing and how do you think this will be affected by the new IMO regulation?

[Fabio Schvartsman, executive]

Just to correct a little bit. We are not convincing them. Actually, we're paying for the equipment that is going to be installed in all the ships for this purpose. So the -- as we are paying, it means that we are getting a lot of traction here. And we consider this point that we have most of our fleet converted already for the scrubbers. That means that we are poised to actually benefit from the reduction in the costs of the bunker that is more quality or more productive. And therefore, we are good here. Exactly as we have mentioned in the last calls with you.

[Operator]

Our next question comes from John Brandt with HSBC.

[Jonathan Brandt, analyst]

First, I wanted to ask you about, I guess, your shares. Firstly, on the share buyback. As you said, you've got about 50%. And I can see that they're still sitting in treasury shares. Are there any plans to actually cancel those? Or should we expect to see them as treasury shares going forward? And related to that, I see that BNDES has been selling down part of their stake. I know it's not really your decision, but I was wondering if there's any update -- if you have any update as to whether or not the controlling shareholders might be looking to sIt this all down. And then, I guess my second question was related to the iron ore, the pricing and the premiums. So you said about 79% of your volumes that were sold are premium products. But my understanding is, you're not actually getting a premium price for all of those products because of the contracts. So I'm wondering if you can sort of shed some light on how much of your volumes are being sold by contract? I'm really trying to get a sense of, if iron ore prices stay the same because of the mix shift going from contracts to using the new Metal Bulletin benchmark that you've created, how that will impact premiums? How much higher they can go?

[Fabio Schvartsman, executive]

Okay. Let's start with the share buyback. If I understood properly, you want to know if we are going to continue to buy back in the same speed that we've done so far. Was that your question?

[Jonathan Brandt, analyst]

No. I guess I was wondering if you were supposed to cancel the shares that you bought back or maybe clear in your balance?

[Fabio Schvartsman, executive]

Yes. The idea will be to eventually cancel that. So we have no other plans for these shares other than canceling that. Regarding BNDES and their consolidated roles, the only thing that I know is what I hear from them. And what I hear from them, especially from pension funds, is that they are not willing to sell in the short term. So I have no idea if, on the other hand, BNDES is aiming to sell or not to sell. It's something that, unfortunately, I don't have information. Now regarding iron ore prices, I will pass to Luiz Meriz to give you more color in this issue of contract that you're asking.

[Luiz Meriz, executive]

Alex, thank you for your question. Basically, the premium, when you're capturing a rise from 3 major families of products, let's say, right, you can say that the pellets, which have their prices discussed on a yearly basis, the Carajás, which is already fully indexed to the index, which represents pretty much the value of the Carajás. And then the third part of that will be related to the blend, which is a lower alumina price. But the majority of this blend is sold on a yearly basis. So we expect as from the next quarter, you already will be in a position to capture the value that the market is recognizing on lower alumina quality on our products.

[Operator]

The next question comes from Alex Hacking with Citi.

[Alexander Hacking, analyst]

On the first question, I just wanted to follow-up on the previous answer. Is the goal to sell 100% of Brazil Blend Fines linked to the lower alumina index next year? And then, I guess, second question just on Moatize. As the mine ramps up, what's going to happen to the mix of met coal and thermal coal? Because obviously, I think there's been a bit more thermal coal in the mix than what was originally anticipated. I'm just trying to figure out like steady state once the mines at full capacity, what does it look like?

[Luiz Meriz, executive]

Alex, thank you. On the first query related to the booked deck, I mean, contracts have a different duration. I mean, the majority of those contracts are done on a yearly basis, as I just mentioned. So on those ones, this will be implemented in a faster rate. For some others, you may have a small part of that, which obviously is slightly longer and then it will take a few more time to happen. In general, the market recognizes that the alumina is a natural movement in terms of pricing. They have a specific index now, which is being published, which represents the lower alumina on our simple feeds and is the most fair way to assess the value of this product.

[Luciano Pires, executive]

Alex, in terms of Moatize, the mix. We have several different pits and sections on the mine. The current sections we are mining, you're correct. So it has a slightly lower proportion of met coal. The new sections, which we're going to be opening as of now and next year are much richer. So, therefore, the situation in '19 will be similar in terms of volumes will improve surely in the second half as I pointed out, but the mix will continue to be similar. In 2020, the mix improves a little further. And in 2021, it improves dramatically. So the profitability of Moatize will increase over time but get to the big, big jump from 2020 to 2021.

[Operator]

The next question comes from Christian Georges with Societe Generale.

[Christian Georges, analyst]

My first question is on the pellets. Could you give us an idea of what is the breakdown between your pellets or how much of your pellets do you send to China these days compared to what -- I guess, you send more to Europe? And linking to the contract question earlier, did you hear some contracts in pellets, which are maybe 6 months near on the European destinations?

[Luiz Meriz, executive]

Yes. Christian, thank you for your question. Well, China is not traditionally a big consumer market for our pellets. I would say that the majority we will stay in Europe and Brazil. So that's -- basically, is I would say, about eventually less than 10% of our volumes are directed to China. But our full production is committed in the long-term contracts. That's the scenario.

[Christian Georges, analyst]

But the proportion in China is increasing, right?

[Luiz Meriz, executive]

Not necessarily right. We are increasing -- this year, eventually, we might see a small amount going -- heading to China as the production increases. But that will be -- eventually sprout sales, right? The majority of the volume is in long-term contracts, of which China is a relatively small part of our competitor.

[Christian Georges, analyst]

And then my second question on annual fees. So we're looking at 17 million tons in 2020. What is the target for 2019? I mean, are we getting to be flat throughout the rest of 2018 and gradually going onto 2020 at 17 million tons? Or is it just linear?

[Luciano Pires, executive]

We will provide the full guidance by Vale Day, so I just wanted to give you a direction of where we're heading. So we should be at the rate of 17 million tons for the second half but were still sorting out the details so we'll give you precise guidance for the full year '19 at Vale Day.

[Christian Georges, analyst]

Okay. And very last thing, VNC was -- back industries. Does this -- nickel linear, is this -- is that a disappointment to you? Or is it part of the plan that you currently have?

[Fabio Schvartsman, executive]

Look. We never hided the fact that VNC is a very difficult operation that Vale has, and we have a new -- the new management of these metals are very focused on changing that for better. So we are now in this process. We want to reach a decision of what is that we are going to do if VNC shortened. And as soon as we have this decision, we are going to share in the market.

[Operator]

The next question comes from Grant Sporre with Macquarie.

[Grant Sporre, analyst]

My first question is just a little bit more on VNC, if you're able to share anything. The question really is around if you think you've -- it's more of a revenue problem than actually a cost problem at VNC in terms of some of the realizations you might be getting from the intermediate products. So I don't know if you could share any details on that. That would be my first question. And then the second one is that one on Moatize. You guys have guidance at Vale Day '18 in terms of how the cost would evolve. Is that still your current thinking as to how those costs would evolve over time?

[Fabio Schvartsman, executive]

Regarding VNC. VNC is not only an issue of price, it's an issue of cost as well and it is an issue of capacity. We are operating at low capacity utilization at this point and we are focused in increasing that. If we can increase that, we are going to save thousands of dollars per ton that will be as important as price increase. So there is -- there are 2 issues. The future prices, obviously, eventually can help, but we are not counting on prices. We are looking if we are able to be in a position to produce a lot more from the same equipment that we have there. Luciano, on Moatize?

[Luciano Pires, executive]

On Moatize, we have no change to the cost guidance that we provided at Vale Day. And just reminding everyone again, we look forward in the longer term to have a total cost structure in Moatize at around $80 per ton cost at port, which means $60 per ton of operating cost plus $20 per ton of an additional tariff required to service the project finance. So very competitive. Obviously, it will take a few years to get there.

[Operator]

The next question comes from Tyler Broda with RBC.

[Tyler Broda, analyst]

I just wanted to say -- I just want to ask, on Salobo, the third concentrator coming in. Is that going to be the same size and design as the previous 2,000 trailers? Or is there any changes? Or what we should be looking for in terms of any specifics on what that might change in terms of the cost profile. And then -- and secondly on that, maybe you could just add a bit more clarity exactly on what needs to happen before the Wheaton money comes through? And then just secondly, on Samarco, the pellet market is still quite strong, just if you could give an update on where things are there at the moment.

[Luciano Pires, executive]

Okay. Tyler, on Salobo, the -- it's -- this expansion is exactly equal. It's a sister plant of Salobo I and II. So in theory, it should add 50% of capacity. But the reason why Salobo doesn't go from 200,000 to 300,000 is because copper grades will decline over time, so therefore the peak production will be, actually, 268,000. So when you think longer term, there is then a trend for Salobo cost to come slowly up, although from a very low level. You all know that we have cost there, close or below $1,000 per ton. But on the other hand, on the first few years of the startup of Salobo III, actually, there is a counter-effect that keeps costs at the same level, which is the fact that there's an already mine stockpile of ore, very close to where the new crusher will be for Salobo III. So transportation businesses will be 0, mining cost will be 0. So expect Salobo cost to be stable at least over the next 5 years.

[Fabio Schvartsman, executive]

Well, regarding Samarco this year. This year was a year of important achievements in Samarco. First, we had an agreement with all the parties involved, especially the prosecutors. That was a very complicated thing to get. And this was the most important step forward because it's taking out the uncertainty of which will be the impact of the liabilities that we had because of the accident, that Samarco has because of the accident. The second good news is that very recently we decide to start the construction of the tailings dam that will support the restart of the operation, called Alegria Sul, where we not only the site will start but we got the licenses to build it. And we are under construction right now. That means that by the end of 2019, we will have everything in place to ask for the license for restarting operation. Therefore, if we are able to get these licenses, we will be -- we will have everything ready for restarting the operation by the beginning of 2020.

[Operator]

Our next question comes from John Tumazos with John Tumazos Very Independent Research.

[John Tumazos, analyst]

Two questions. Should we expect another cobalt stream in New Caledonia to finance the next tailings project there? Second, I was rereading my notes from the October 2011 Mozambique Vale tour. And the guidance was 77% net coal, trending to 22 million tons future output, fully developed. There was emphasis on the Chipanga seam, 25 meters thick, but the other seams were 17 or 10 meters or smaller and variable. Is there a difference where, when you open up and get into the mine, it just isn't as good as when promised?

[Fabio Schvartsman, executive]

John, this is Fabio. Let me answer the cobalt stream, and then I will gladly pass on the seam because I was not there, and so I have no idea what was promised back then.

[John Tumazos, analyst]

I will send you my notes, Fabio. I already started to write you a letter.

[Fabio Schvartsman, executive]

Thank you. Regarding cobalt stream. I would like you to understand that in our point of view it is a totally different story, our cobalt stream in Voisey's Bay from the one in VNC. The reason is a very simple one. In Voisey's Bay, we didn't have the cobalt. So we -- the cobalt is totally dependent upon opening the mine. If we didn't open the mine, no cobalt would be available, therefore we sold something that we didn't have in order to make the mine feasible. That was the proposal. VNC is a totally different story. We already have the cobalt in our stream of revenues and the cobalt is there. So this -- the impact of streaming in New Caledonia will be neutral. So the benefit that we have in the case of Voisey's Bay would not be present in the case of VNC. I hope this answers your question.

[Luciano Pires, executive]

So John, on Moatize, starting by the production volumes. So the 22 million tons encompass 18 million tons through the Nacala Corridor plus 4 million tons through the Beira Corridor. So as you know, we ceased operations in the Beira Corridor so we would, theoretically, be limited to 18 million tons. However, we're going to go to 20 million tons based on debottlenecking of the Logistics Corridor. So what's going to limit the capacity, currently, is the logistics. In terms of the split, the 77% will not be achieved over life of mine. Now we know more about the ore body and it will probably peak once we go to the new sections that I mentioned, section 5 and section 6, 65%, from today's breakdown of 55%. Chipanga will not be as meaningful as it was before. So the quantity of Chipanga is reduced. However, the other seams of coal that we are mining, they have also very good quality. They have strange names like Sousa Pinto and Novo, whatever, but we are achieving very good price realization. So the fact that Chipanga is decreasing should not be a concern in that respect.

[Operator]

The next question comes from Alfonso Salazar with Scotiabank.

[Alfonso Salazar, analyst]

The first question -- the first one is regarding the payment that you will get from Wheaton once Salobo III is up and running, if you can provide more details or anything else that you can share about this. The second is regarding the premium that you'll get from -- for the quality of your premium in iron ore. How do you see that in the long term? And what will you see, especially in the supply side, if you think that outside Australia there could be more supply of high-quality ores that will compete with Vale's? And how do you expect that to evolve as China uses more scrap in -- again, in the longer term?

[Luciano Pires, executive]

Okay. On Salobo III, the bonus is a function of 3 variables. The first one is the timing of the completion and ramp-up of Salobo III. So the guidance that we gave you, from $600 million to $700 million, assumes that we achieve those milestones in 2023, although the start-up of the project is scheduled for 2021. So however -- so therefore, we have a full year to achieve the production targets. And as I indicated, because it's a sister plant of the 2 which are already operating, that should not be difficult. So timing is one variable. The second variable is capacity. Again, shouldn't be an issue because we've been running sister plants for a while now. And the third variable is ore grades, which, as much as we continue to mine and to know Salobo, there is no surprise that we expect it's being -- that the ore grades have been behaving according to the mine plans. So level of risk for receiving that bonus is very small.

[Fabio Schvartsman, executive]

Regarding quality premium in the iron ore universe. Well, we certainly think that is a structural thing, therefore, permanent. And there is a fact that I would like to emphasize that helps understanding what's going on in the market these days. If you notice in the last few days, the index, Platts, went up in a very important manner. We are now in the neighborhood of $76 per ton. That was not achieved for a long period of time. That was a combination of 2 things: a strong demand in China, especially because of the latest stimulus that the government is making to speed up the economy a little bit; but more importantly, our competitors made the right decision of changing the quality of the product that they are offering to the market. It always come with a consequence. It means that, certainly, there are more costs associated with the decision. But the truth is that now there is more high-quality blended ore coming to the market through the index. And accordingly, the index is moving forward. And this is the best showcase that we can have, that this trend is so real and so important that everybody is now having to adapt to it. Either people are offering more quality products if they have them or they are cutting production accordingly because the market is punishing very much the low quality. So it is implied in this comment that we are quite comfortable that this will continue like that and we don't see anything meaningful coming in the next few years as further supply. And all the initiatives that you eventually heard of are very complicated, very high CapEx, very high OpEx, and there is a clear doubt if these projects will be developed or not. This is the situation as we see today.

[Operator]

The next question comes from Mr. Alex Hacking with Citi.

[Alexander Hacking, analyst]

I just wanted to follow up on the nickel volumes. You mentioned that those would be back to 310,000 tons by 2020, if I heard correctly in the prepared remarks. Could you maybe disclose where that additional sort of 50,000 tons -- at which mines that additional 50,000 tons of nickel is planned to come from? Because if I remember it correct, the Voisey's Bay underground doesn't start until 2021. So maybe just some color there would be helpful.

[Luiz Meriz, executive]

Okay, Alex. Just to clarify that. This is an [indiscernible] to our capacity stalled. So it would be around '21 or '22. Where it comes from is the mines, of course. They have to be sustained on the level -- and we have a good example of Voisey's Bay. So long harbor for you because we have spare capacity of 10,000 tons there. We still have capacity in Sudbury of over 50,000 as well. They have created real capacity. And as Fabio mentioned, we have a huge increasing capacity in New Caledonia for nothing, just for better management and small debottlenecks that we have to do there. So there are a lot of capacity on our industrial plants to get to that. So the best to get to 310,000 will be guided by the market, by the way. So as we expect, the crossing of the curve is 2020. But I would say, this guidance -- we made more clarity on Vale Day, by the way. It's just a direction, okay. Just to bear in mind that we are able to produce -- the mining investments, just to get it, it's very transparent. It's like the second, first, and also third, but that's very marginal. But that's, I guess, a lot of volume as well. But anyhow -- anyway, the direction is 310,000. It's not -- and it's going to be achieved as market evolves very, very cautiously and, as we mentioned before, with a better cost structure because we're going to keep our -- actually, we're going to reduce our fixed base to run that whole operation. I hope it clarified you.

[Operator]

The next question comes from Thiago Lofiego with Bradesco BBI.

[Thiago Lofiego, analyst]

Fabio, I have one follow-up question. Could you give us an update on the railroad concession renewals of Carajás and Vitoria-Minas? What's the timing expected for those renewals? And what is the potential CapEx linked to those?

[Fabio Schvartsman, executive]

I will start, and then I'll let my colleagues here complement that. But the state of the art here is very best. We are the agency see right now. We are just waiting for the final decision of the agency. I think that by cautious reasons, the agency, it is waiting for the change of -- or the elections, the change of the president, in order to move forward with this. But it's -- it will be a low-hanging fruit, waiting for a new president to catch. So I'm pretty optimistic that as soon as he starts in the office, he will be allowing this process to move forward. We have everything in place to deliver this as soon as the executive -- Brazilian executive power decides so. Regarding the investment, I don't have the numbers and how they are going to be.

[Unknown Executive]

Well, I think it's quite soon to say a number because in the agency, as Mr. Schvartsman said, we're going through a very technical analysis. It's not final yet. And after that, it's going to go to the union court for the final analysis, and then back to the government for final decision, and then the renewal of the contract. So up to now, I would say that it will be early to call a final number because we are going through a technical methodology that is about to be concluded by the agents.

[Fabio Schvartsman, executive]

Allow me to comment on that. If you take, for instance, the Chico railroad that is selected by the government to become the counterpart of the -- for the concession. This will be dealt by Vale once the renewal of the concession is approved. But it will take several years to build it. Therefore, the investment no matter how much it's going to be it will be spread among many years.

[Luciano Pires, executive]

And this is Luciano complementing. This is a general cargo railway, which, by nature, costs much less than a heavy-haul railway like and iron ore railway. So if you take the numbers from the iron ore industry, they are not a benchmark for this railway.

[Operator]

The next question comes from Marcos Assumpção with Itaú BBA.

[Marcos Assumpção, analyst]

First question on production and sales volumes in iron ore. In the first 9 months of the year, Vale sold 95% of its production volumes, basically to support its strategy of increasing blended volumes. How should we expect that pressure to behave in 2019? And the second question, if you could comment a bit or give us an update on the pellet premium negotiations for next year as well, at least, how the market is so far. Supply demand is behaving in the trend that we -- you see for pellet premiums for next year.

[Fabio Schvartsman, executive]

Thank you, Marcos, for your question. Regarding the 95% sold last year, most likely we will have the same situation this year, in 2019. And the reason for that is because we are using our flexibility. And as we -- today, we are in a different position from other companies as we have a lot of inventories sitting in China. We can speed up sales or slow down sales according to the behavior of the market. As naturally the market slows down by the end of the year, so naturally we are going to most likely hold a little bit of sales for this -- for next year. And this explains the level of sales in proportion to production that we had in the first quarter. Besides that, we are coming to a more close comparison between what is produced and saved every quarter as we, now, we are coming to the end of the building of inventories in China.

[Luiz Meriz, executive]

Yes, Marcos, regarding the pellets. We basically are starting our negotiation season, right? I mean, the market balance, the showing up of the increase on price. But we rather not be much more specific than that due to the sensitivity of the negotiation moment that we are in now.

[Fabio Schvartsman, executive]

But the truth of the fact is that we are completely oversold. And consequently -- and the demand is higher than our capacity. So the natural consequence that there will be some price adjustments because of this unbalanced situation that I just explained.

[Operator]

This concludes today's question-and-answer session. Mr. Fabio Schvartsman, at this time, you may proceed with your closing statements, sir.

[Fabio Schvartsman, executive]

Very good. Again, as always, it was a pleasure to have you on this call. Very detailed questions that were presented. And we are very pleased to tell you that Vale is moving forward in every aspect that we are looking for. So it seems that we are poised to have another good quarter in front of us. And I hope to have you all in the next call by the beginning of next year. Thank you so much, and have a good day.

[Operator]

That does conclude Vale's conference call for today. Thank you very much for your participation. You may now disconnect. Have a good day.