ArcelorMittal SA (MT) 2017 Q2 法說會逐字稿

Please go ahead, Daniel.

Thank you. Good morning and good afternoon, everybody. This is Daniel Fairclough from the ArcelorMittal Investor Relations team. Thank you very much for joining us on this call today, which for your information is being recorded. This morning we published our results for the second quarter and first half of 2017. We also published a presentation of these results together with our strategic progress and this was accompanied with detailed speaker notes and a Q&A document as well.

So we hope that you've had a good chance to review these documents and so we intend to move directly to your questions on this call today. Given the call should last about 1 hour. (Operator Instructions) And with that, I will hand over to Mr. Mittal for some introductory remarks.

Thank you, Daniel. Good morning, good afternoon, everyone. Thank you for joining today's call to discuss ArcelorMittal's first half 2017 results. I am joined today by Aditya Mittal, Group CFO and CEO, Europe; Simon Wandke, CEO, Mining; Genuíno, our Group Head of Finance; and Daniel, Head of Investor Relations. Before we start our Q&A session, I want to provide some opening comments on the results and our outlook for the remainder of 2017.

Starting with safety. The injury frequency rate in the first half of 2017 was essentially stable compared with last year. However, we continue to prioritize further improvement in our safety performance and strive for 0 harm.

Our financial results in the first half of 2017 were the best since the first half of 2012. Group EBITDA increased by 61% year-on-year and we delivered twofold increase in net income and a healthy return on equity. This reflects not only the improved market backdrop but also the ongoing benefits from our unique action 2020 plan. Given the improved macroeconomic environment, we have today increased our forecast for global steel demand by 200 basis points.

The strong demand backdrop is helping to support robust steel spreads. Aside from our stronger financial performance, we have made progress on a number of strategic fronts. First, our balance sheet continues to strengthen. At 1.5x, our net debt to trailing 12 months EBITDA is at its lowest level since 2009. This will improve further over the remainder of the year as we continue to make progress towards an investment-grade credit rating.

While further deleveraging remains a priority, we are also capitalizing on growth opportunities. This includes our investments to support our continuous shift towards high added-value products as well as the recently announced acquisition of Ilva in Italy. Ilva is an exciting opportunity to create value for shareholders. Contrary to perceptions, this is a Tier 1 asset. It has a scale, a deep water port and very high quality finishing operations.

That we can add this asset to our portfolio without compromising our balance sheet is a great news for our shareholders. Finally, we continue to make good progress on Action 2020. I think many of you will have attended our recent site visits to Gent, where you had opportunity to see firsthand the ongoing transformation that is underway. We are now operating from a more efficient resized footprint and we are utilizing the latest technologies to enhance operations, to drive productivity improvements and support maintenance excellence.

So to finish, the strengthening market backdrop suggests that shipments in the second half of 2017 will be at or above first half levels.

Together with healthy steel spreads, this provides a supportive outlook for the second half of the year.

With this, now I'm happy to take your questions with my team.

So we'll take the first question please from Mike Shillaker at Crédit Suisse.

So my first question in terms of the outlook, it all sounds pretty rosy. But can you give us a little bit more color because there's obviously a lot of moving parts with seasonal volumes on one hand, spreads should remain very good. But if you're actually talking about volumes coming in, in line or actually potentially even better H2, H1, does that mean we ignore seasonality completely? And given where spreads have moved through the year, does that mean that H2 actually could be at least as good if not better than the first half? And within that context, a sub-question. Are there any areas that you're particularly worried about? And certainly, could we have some comments on U.S. also, because that has been something that has come up on a number of conference calls, I think recently in terms of the risk to the demand from the U.S. auto sector. And my second question, a bigger picture question really is, if you look at what's happening in the world of resources at the moment, companies do seem to be getting rewarded for deleveraging. Aperam was a great example, but now you've got Rio Anglo focusing very much on balance sheet and getting debt down similar before they move to shareholder returns. Is there any reason why you wouldn't now follow the same approach post-Ilva and focus 100% on deleveraging? Unless, of course, another asset like Ilva came up. But in the absence of that, would you focus completely on balance sheet deleveraging? And also within that context, are there any assets that you have that you think could be nonstrategic? And I'm thinking maybe Erdemir or China Oriental or something similar that you could think of disposing to actually accelerate that deleveraging process?

Okay, great. Michael, thank you for all of your questions. I think you've asked a lot. But let's try and get through some of them and then if you want, feel free to ask in more detail. So just in terms of outlook and spreads, I think we have improved our outlook for global apparent steel growth 2017 by 2%. If you look at the breakdown, this is more a China effect and a CIS effect. We have actually reduced outlook for the U.S. and moved to the upper end range of our European outlook. Brazil is also down actually. So outlook is better, which points to a better steel environment and that's how I would read it. I would not read it as a better outlook in terms of underlying core demand. Our core demand still remains very healthy. PMI is at a 5-year high. And when I refer to PMI, I'm talking about the ArcelorMittal weighted PMI based on the markets in which we operate. So that's just in terms of -- the outlook in terms of steel consumption globally. In terms of volumes, we're seeing that the second half would be equal to or better than the first half. So I'm not suggesting that it's clear that the second half would be better than the first half, but equal to or better. So slightly better is a much better way of paraphrasing our remarks. In terms of regions, you talked about dismissing seasonality. I've not dismissed seasonality. There still is a seasonal effect. But when you look at our first half shipments, they were relatively weaker than what they should have been. And the weakness is primarily in long products. It was in Brazil, it was in Ukraine, because we had a blast furnace outage. In terms of Europe, we had some weakness in some of the higher added-value products in long. So as a specific product, rail is a good example. So we expect some of that to reverse in the second half. So what does all of this mean? Roughly NAFTA is flat. Europe, flat. Steel be down, but long steel will be up. So overall roughly flat as well. ACIS will be up for the reasons I just spoke about and so will Brazil. That's our rough expectation when we provide you with this framework. In terms of spreads, I think you can calculate spreads yourself. I will not comment. And you know how spreads are changing, first half versus second half. What's happening to contracts lag as well as the cost of raw materials. In terms of U.S. auto, yes, demand has weakened, but the overall level is still quite healthy and sustainable. And I would expect that in the medium term, even if -- this demand level is still very good for the steel industry. It's much higher than what we've seen in the past and it's not something that I'm overly concerned about. In terms of focus on deleveraging, you're right. We are focused on deleveraging. We believe it is good. We believe it is the right decision for the company. We think we end up lowering our cash requirements, lowering our cost to become that much more competitive. And what really ends up happening is the percentage of EBITDA to free cash flow rises. I think we see some of that effects already now, where we have lowered our cash requirements from $5 billion for the year to $4.6 billion, which implies EBITDA to free cash flow has just moved up by that amount. I'm not going to comment on our specific investments. We like the investments that we have. And so I hope I've answered all your questions.

Yes, is there any sort of mix reasons? Because if volume is at least as good H2 and we can calculate spreads, then we can come up with our own assumption as to whether H2 is as good or better than H1. But is there any mix issue, either product mix or region mix that we may not think about, that could mean that H2 would be weaker than H1?

You're trying to box me into guidance, Michael. I don't think we have said H2 will be better or worse than H1 or equal to in terms of EBITDA. So the only discussion is on volumes. In terms of mix effects, I don't see any significant mix effect into H2 versus H1. I think Brazil will export the same amount, maybe slightly lower as the domestic market recovers. But that would be the only area that I would point out. I think the rest, hopefully we have the pickup in HAV and long in Europe. HAV generally is fine, but just in long, we had weakness in rail, so we need a good pick up there. But other than the effects reversing first half into second half, I'm not seeing any big mix changes.

So we'll move to the next question, please, from Ioannis at RBC.

Two questions on my side. First on the cash requirements. When you originally guided to $1 billion worth of capital investment in February, iron ore was at the $90, now it's $20 lower. And so my question here is twofold. The incremental $500 million of working capital investment, is it exclusively due to better volumes in the second half? Or is there anything else included? And also have you changed your implicit iron ore price forecast to arrive at the working capital guidance? And then secondly, if I just want to ask you on the FX movements we've seen recently. A recent trend is the appreciation of the euro. And in the short term, I guess that has a positive FX translation impact from your European earnings. But if this trend persists, could it lead to some loss of competitiveness for the European steel operations and also some of your end-users like the automotive sector? Because we are now at that door of '17 and if that trend continues, just want to hear your thoughts on potential mitigation and measures that you can take to offset any loss in competitiveness?

So in terms of working capital, we are seeing a net buildup of $1.5 billion for the year. And you're right, that is higher than what we had outlined a few months ago. I think overall volumes are better, but overall the price level, I think even if you look at iron ore, it is not as high as it was a few months ago, but it is still higher than what you would expect as the long-term average for iron ore. So I think that some of that is reflected in our revised working capital guidance. I would also add that this is an estimate and clearly the investment could change based on changes of prices for steel or prices for raw materials. But this is roughly our expectation for the remainder of 2017. In terms of earnings and the euro effect, I think you're right. At the end of the day, a strengthening euro has positives and negatives. So let me just address positives. Normally a currency strengthens because demand environment is good. It demonstrates that the macro fundamentals are strong. And so from that perspective, I'm not concerned, because a stronger euro is also a reflection that the Eurozone economies are doing better. Clearly, the negative effect is what you pointed out, that Europe relative to the rest of the world has higher cost. And I think all companies which are based in Europe including us are focused on that. So we have, at least at ArcelorMittal, very clear plans on improving productivity, reducing our cost and clearly those pressures only increase when the euro is strengthening. I would imagine our customer base, which is automotive, would also have similar plans. So that is how we would mitigate, but I think a strong euro, at the end of the day, is not only negative news.

And just a follow-up if I may, just on the taxes and pension payments, but you're guiding them lower now. Given the strong earnings outlook, I'm a bit surprised that you're guiding to lower cash taxes. Could you please elaborate on that? And also on the pensions, what's driving the lower payments?

So the taxes is a forecast. And really what is happening is we have more income in countries where we have higher net operating losses. And so the cash tax effect, as a result, is less. So I don't know if that's clear to you, but we have -- in some countries, we don't have any NOLs because of the profits we have made over the last 5 years. In other countries we do. And the increase in earning is more weighted towards the countries in which we have higher NOLs. In terms of pension payments and others, I think this is a combination of rates, the growth in terms of our cost as well as better management of the plan and changes on the discount rate. Did that answer your question?

It does.

Sorry, we could not hear you.

That answers my questions.

Thank you, Ioannis. And we'll move to Alain at Morgan Stanley, please.

Just one question from my end. Do you mind commenting a bit on the Aperam demand in Europe and the risk of potential destock in the second half? It looks like the apparent demand was very strong in the first half. Where do inventories stand? And where do you see the risk there?

So I'm not sure where you see apparent demand very strong in Europe. My number for apparent demand in Europe is 1.2% growth versus the first half '17 versus first half '16. So perhaps we're looking at slightly different data points. As I pointed out, we are now for the year at the top end of our range. So our range for Europe is 0.5% to 1.5%. So the top end implies 1.5% for the year.

So we'll move to Seth at Jefferies, please.

I've 2 questions. First on a top-end question for China and second in your Brazilian business. With regards to China, I was wondering if you could give us a sense of your confidence in sustainability of current margin strength. I think it was roughly a year ago, you flagged in one of these calls the risk of the Chinese steel (inaudible) over earning and the need for some normalization. What's your degree of confidence that current Chinese strength is today more sustainable? And then separately on the Brazilian market, I was wondering if you can give us an update on any signs of domestic demand recovery in the region? And then also from a product mix perspective, Q2, you shifted towards more lower margin exports. Do you expect this mix drag to continue into H2? And where do we expect the Brazilian market to go over that time frame?

Okay. In terms of China, so just talking about steel spreads for a moment. As you know, our range of steel spreads, this is hot rolled price minus the cost of raw materials, is between $130 to $170. If you look at the first half of China, it was at $160. If you look at where spot spreads are, as you pointed out, they are higher than this range. So clearly we do see downside risk to this level. How would the downside risk manifest itself? I think if Chinese growth slows, then we have a downside risk. So far what we're seeing is that China is actually outperforming. We see that in terms of the real estate sector as well as the machinery sector, but clearly the downside risk exists. In terms of what does that mean for ArcelorMittal, I would not say there's a direct translation of this spot spread into our markets. I think the $160 that existed in the first half reflects the spreads that we had in the first half. Today, if you look at just spot spreads, for example, the delta between Southern European prices and China are much lower than they had been in the past as well. So our end markets have not really reflected this increase in Chinese spreads. In terms of Brazil, I'm going to get Genuíno to answer the question.

Yes, so in terms of Brazil, we see already a good development in second quarter versus first quarter in terms of higher shipments domestically. So this is sure both for flat and longs. Flat even better, because you also see growth year-on-year and you see quarter-on-quarter. Longs, you see an improvement quarter-to-quarter, to some extent of the seasonal. Year-on-year, we are still down primarily because of construction. So construction is weaker than what we had initially anticipated. And one of the reasons why we have been down on apparent steel consumption demand for Brazil. So going forward that we would expect trend should continue. We would expect the domestic shipments to continue to improve. So it's a positive development from that angle.

If I could just ask one follow-up question with regards to China. You highlighted the downside risk coming from the demand side. Do you give much credibility to the supply side reform and capacity closures, both blast furnace and induction furnace we've seen year-to-date? Or is that still not enough to offset where you see utilization rates in the region?

Yes, that's a very good question. From our point of view, China has 300 million tons of overcapacity. Year-to-date, there's about 105 million tons of capacity that has been shuttered. We think that they should shutter another 200 million tons. That achieves a high level of capacity utilization, so they still have some way to go in achieving those numbers. Their number is about 150 million, so according to the Chinese, they have about 45 million tons to go. The delta between our number and their number could just be that they have a much higher growth forecast, i.e. demand forecast for Chinese domestic production. If so, maybe the right number is something in between. Your question was more how does that impact spreads. At this point in time based on the capacities that they have taken out, 105 million, we have not revised our range of 130 to 170.

Thanks, Seth. So we'll take the next question, please, from Alessandro at Berenberg.

Just have 2 questions. The first one is related to what's happening at the moment with the massive reduction of exports from China, which is clearly impacting more probably long steel, considering the decline of capacity in the induction furnaces. I was just taking a look at the output in terms of production of your long steel in Europe, which has not really dramatically changed since Q1 2012. How do you see in terms of knock-on effect or the continuation of lower export from China in terms of potential uplift on margin on the long steel in Europe, even though there is not really a visible improvement in the level of demand for construction at the moment? And the second one is clearly related to a kind of phenomenon that has materialized and especially in the Russia in the last few months. When we've seen steel prices up significantly from the Black Sea despite the relatively weak ruble relative to both dollars and the U.S. If we assume that China remains relatively stable, so I'm not really pricing any kind of improvement, maybe decline slightly from the current level, utilization rates should continue to sustain margin in China, considering the shutdown. If the situation remains at the moment the way that you've said, considering utilization rates in long steel in Europe, do you see a potential uplift on margin because of lack of competition?

Alessandro, thank you for your question. In terms of the long business in Europe, I think 2 comments. First, let's look at our business in Europe and you refer to Q1 production levels today. There has been a massive shift in our operating footprint. And so the numbers that you see don't necessarily reflect that. Since 2012, we have shut a number of assets. We have sold facilities. For example, last year we sold Zaragoza, last year itself we shut our Zumárraga facility. We have also shut facilities in Madrid as well as in Luxembourg and other parts of Europe. So the production level is reflecting higher output from the remaining assets. And we've also moved our production into more higher added value segments. So rail I spoke about, but we're also doing a lot more in terms of coiled wire rod, sheet pile and sections which traditionally have higher margins and even today have higher margins on the commodity side. So that's just on our European long business and what has happened over the last 4 years very quickly. In terms of margin for long steel Europe and the correlation with Chinese exports, I don't see it so clearly. Because in Europe, we still have significant overcapacity in the long business, especially in the construction sector and those segments. That's why consciously over the last 4 years, we have changed our order book and we have changed our asset mix. Clearly to the extent that construction recovers, today we can see that things are looking better for the Eurozone. That will be a strong positive for our European business. And clearly, the last thing is that the margin for long in Europe is also very dependent on scrap movement. So you'd have to model that in as well as the growth in construction to really determine margin. I didn't really understand your question on Black Sea pricing. So perhaps (inaudible) understood it so maybe you can go ahead and -- sorry?

I think you already answered the question, Aditya.

Okay.

But maybe, Alessandro, you can correct me, but I think (inaudible).

Yes, just a very little follow-up because we've seen the steel prices, of course, picked up significantly in the last couple of months, despite the weakness of the ruble, which is quite atypical. So Russia is not really fully utilizational yet. There is an uptick in domestic demand, which is visible. But I was very much surprised by the fact that steel price have picked up so much despite the weakness of the ruble. Clearly, the incentive for Russian producer was for more leverage around the fact that translation wise is beneficial for them. I was wondering whether you have any kind of macro picture that justifies this uptick.

Nothing specific, apart from the fact that demand in Russia is better. We have also revised upwards our apparent steel forecast for the CIS market. So that clearly has a role to play.

Thanks, Alessandro. So we'll take the next question, please from Cedar at Bank of America.

Two questions. Can you please talk about the South African business? I see that it's struggling a little bit because of weak demand, particularly in the long space. And I note in the release today on the SA side, there was an impairment taken there and the management team talking about potentially looking at structural initiatives or structural changes. What can we expect to see from the SA business? And then if you look at the U.S. market, we've obviously had a few price hikes announced over the last few months, which appear to be sticking. Can you talk about how you see the outlook for prices in terms of further price hikes potentially coming through considering that ultimately global pricing is healthy, but you do flag that the autos market, while still at a healthy level, is maybe rolling over a little bit and you still have very high imports into the U.S. market at the moment? That would be helpful.

So thank you for your question, Cedar. I'm going to get Genuíno to discuss the impairment that we took in South Africa. Very quickly I'll answer the macro question. Our South African business has been hit by 2 macro headwinds, which actually normally should work in the opposite direction, but not in South Africa. The first is that we have a weak economy, which means we have weak demand. And at the same time, the currency has strengthened. And that has really made the business less cost competitive than it needs to be. And it has the inability to maximize output because there is lack of demand. And then that is causing stress in business and causing negative EBITDA. What we have done as a result is we have a slate of cost reduction programs. We've also made management changes on the operating side, and on the commercial side, we're further strengthening the business. And we hope that the changes that we have outlined will support the business in the medium term. In terms of the North American market, I mean the North American market is still doing well. We're not overly concerned by the weakness in auto. The demand level is still very healthy. We continue to sell more of our advanced products into the U.S. automotive market. So overall, the auto franchise in NAFTA is intact if not improving. Overall demand levels, price levels, I think are healthy. And I don't see significant concern in our NAFTA business.

So Cedar, in terms of the impairment in South Africa, it's really a function of the weaker results in longs compared to our expectations at the beginning of the year. So then you are required to re-perform your impairment tests in certain circumstances, which was the case for the long business. And as we adjusted our expectations also in terms of future cash flows, we were required to report this impairment charge.

Okay. And can I just ask one follow-up, just on SA? We have had some plants downsized and closed last year. When would you take a decision to resize the footprint there further to the extent that the economic landscape didn't really improve anytime soon?

It's a good question, Cedar. At this point in time, we are focused on the existing footprint and really optimizing the existing footprint and reducing the cost base within the existing footprint.

Thank you, Cedar. We'll move to the next question please, from Novid at Cowen.

So first, pretty big adjustment to the apparent steel consumption forecast for China. I was just wondering if you can give us a little bit more of the moving parts there, maybe the potential -- what city tiers that you're seeing the strength in this real estate? Just anything that you can give us to kind of highlight that shift from what's changed essentially.

So in terms of China, we see a much stronger construction environment than what we had anticipated. It's still not growing on a real basis, but it is not as negative as it used to be. The markets which continue to grow is automotive, which is -- we're forecasting roughly 5% growth. Machinery segment, which is about 7% growth and the construction market to remain flat. As a result, in China last year, there was limited growth. We're forecasting that real steel consumption will mirror apparent steel consumption in 2017. Manufacturing PMI has also remained above 50 in China, pointing to growth. Is there anything else that you're looking for or...

And then just with respect to real estate, anything that you can mention there?

So Daniel, do you have any specifics you want to go through here?

Not too many specifics. But I think what we recognized is that the strength is in the Tier 3, Tier 4 cities and where you've seen a significant reduction in the overhang of unsold inventory. So that's been significantly reduced during 2017, much more than we'd anticipated. So as a result, there's more longevity to the construction outlook than we previously anticipated.

When you think about Tier 3 and Tier 4 cities, is there more leverage that you guys see there relative to the number of units, given the lower price of housing? So the potential lever that you have for a similar dollar amount could be multiple times out of, say, a Tier 1 city?

I mean, to be honest, I don't know the answer to that question, Novid. So but what I can do is follow up internally and come back to you with a more considered response.

Sure, that'd be great. And then my last one, just switching gears to the U.S. market. You touched on U.S. pricing. We have seen prices moving higher. How much of that do you guys attribute to raw material costs versus potential tightness from Section 232 we've been hearing that imports have come down a lot. And so maybe later this year we'll be actually seeing that number in the DOC data. If you guys could just comment on that?

So in terms of inputs, inputs have not really come down, right, because inputs up to May are high percent, higher year-on-year. What is pointing to relatively healthier U.S. economy is that we still have strong GDP growth. There is some recovery in the oil and gas sector. Manufacturing is also picking up. Auto, yes, we spoke about auto not growing, but not rolling over. But nevertheless, the other segments continue to do well in the U.S. And so that's offsetting that growth and therefore, ASC is still growing. Also the construction is doing well. So long is growing faster than flat. We expect construction growth to be about 2%. So this is -- I think all of these factors are helping including higher lead times from our business and what we see from the rest of the industry as well as the fact that inventories remain at a low level.

And so you see -- if I understand your answer correctly, you see it more as market driven versus, say, potential trade action driven?

Yes, I think that's a fair conclusion. It's very hard to distill these things and be so categorical. I mean the market is allowing for a higher price level and that's what we're seeing. But that is how we think about it, at least.

Thanks, Novid. So we'll move to the next question please, from Francisco at Banco Sabadell.

I have a couple of questions, please. The first one would be related to your feelings on steel prices in the flat business in Europe, could we see a price increase in the short term, do you think? Also regarding Europe, I would like to ask if there's any specific sector or country which is surprising you in the sense that demand is picking up above your expectations. And the last question would be if you're working with any specific time frame regarding your investment grade recovery?

Okay. In terms of Europe, again, the macro environment remains very constructive. We see strong growth in various sectors; machinery, construction as well as auto, with real demand -- with real steel consumption growth actually slightly higher than apparent, about 2% to 2.5% and apparent ranging from 0% to 1.5%. I would not say it is specific to any region. I think it's quite broad-based across Europe. In terms of investment grade rating, I think if you look at our ratios today, we are in investment-grade territory. We continue to have discussions and dialogue with the rating agencies. And as the macro environment, I think, remains strong or stable, as we continue to perform, they will take appropriate action. They should decide and then take appropriate action.

Okay. And so with regarding the price situation in Europe, I know it's -- do you feel you could -- prices could go up in the short term or...

Yes, so I don't want to comment on specific price. I think, yes. Thank you.

We'll move to the next question, please, from Phil at KeyBanc.

I had a question on the Calvert ramp this year. Has that been impeded at all by the imports that arrived in the U.S. in the first half and/or the existing auto overhang in the market?

No, not at all. Calvert is doing better than it was like-to-like basis as well. So we talked about in our release that Calvert is now about 90% capacity utilization. So clearly the transformation work there has been progressing well. So yes, so no impact of imports or the auto weakness.

So shipments picked up at Calvert in the second quarter versus the first quarter and were a little bit weaker at some of your more legacy operations on the flat side?

That's fair. Our production and shipments in Calvert were up first half '16 versus first half '17. And the weakness in NAFTA, some of it was Mexico, where we shipped less slabs, and some of it was in Dofasco as well.

Okay. And then one more for me. Aditya, if you could talk a little bit about your energy related order book within the U.S. I know you don't do tubular products, but anything you could comment in terms of your substrate demand for line pipe and/or OCTG could be helpful.

Yes, so we're seeing pipe and tube demand rise, actually, so strong growth in 2017 versus 2016. And to the extent we supply into those sectors from Calvert, today we're seeing more demand. But our focus in Calvert remains automotive. And so really that's where most of the work is underway, is ongoing.

Thanks, Phil. So we'll move to the next question please from Carsten at UBS.

All of my questions have been answered.

Okay, thanks, Carsten. So then we'll move to Bastian at Deutsche Bank, please.

Just 2 quick questions left. Just firstly, again on the cash requirement. On the $300 million reduction in cash requirements from taxes and pensions, can you please give us the split between the 2 and also let us know whether the reduction which comes from the pension part could be a bit more sticky into next year? Because I guess the tax part is obviously a bit more volatile depending on your earnings? Then secondly, just getting back on your non-core assets as well, and (inaudible) particular, could you please let us know what your plans are with regards to the residuals take plus the conversion of the mandatory convertible, which is coming up soon?

Okay, thank you for the question. I think at this point in time we have a forecast. So it's very hard to give an exact split. Our forecast between those items of tax and pension is $300 million lower. I would say the larger amount is on pensions and the smaller amount is on taxes. I think the better information or better answer to this is what happens going forward. So going forward, we have talked about, in 2017, the cash requirements of 4.6 excluding CapEx. I would expect that as we continue to delever, our interest expenses would fall. I think as you mentioned, cash taxes are volatile. So I would model in higher cash taxes. And assuming that all things remain equal, I know there are a lot of moving parts in pensions. I would expect that number to not change. Therefore, approximately plus, minus -- plus, not minus. The 4.6 to 4.8 would be a good number for 2018 assuming a CapEx of 2.9. In terms of non-core assets, look, I don't want to make any comments. I think we do have certain investments that are important to us from the (inaudible). China Oriental and (inaudible) are 2 that have been pointed out.

Thanks, Bastian. So we'll move onto the next question from Luc at Exane.

A couple of follow-up. Would it be possible for you to quantify the Calvert impact on NAFTA EBITDA when it comes to Q2? Secondly, when I look at the current market trends, the low level of inventories, the, let's say, high price increase that have been announced. I think it can be argued that the price traction is particular, or course, but may remain shy of the announcement. Do you feel that there is some hesitancy, wait-and-see standstill opinion in the U.S. because of the confusion around Section 232? And therefore, do think that clarity could help price move accordingly?

Okay. In terms of Calvert, you know we don't break out EBITDA by facility or by sector, by units, and so we can't really do that. I think I would just add that the Calvert EBITDA is broadly stable, Q1, Q2. I think clearly, the business has done well in Q2, but it also has higher slab prices coming in. And so that offsets some of the revenue gains in Calvert. In terms of NAFTA, I think we've provided a lot of commentary already. There was a little bit of a wait-and-see attitude that we saw in the second quarter, which has impacted. That can only last for so long. I think there have been statements made that the 232 action is not occurring in the soonest, rather it takes some more time. And then therefore, I would expect the market to normalize.

Can I ask another question on Brazil? Could you maybe elaborate a bit more on the declining profitability Q2 versus Q1? It is mostly long related, given the weakness you are talking and the driver for the weaker demand you are seeing there?

Genuíno, go ahead.

Yes, I think in Brazil, the -- so we had, of course, a mix impact because of the higher shipments to the export market, so that's one impact. So and then clearly also we had higher costs, too, coming from Q1 levels, so it's still impacting our cost position there. And that is really true for both segments. It's flat and long, so the decline is both sides.

Thanks, Luc. So we'll move to the last question, which is from Philip at ABN AMRO.

I have just one left on the interest charges. I was wondering, you're close to investment-grade. If you do get an upgrade to investment-grade, do these outstanding bonds that you have also have a covenant that you -- the interest or the coupons on the bonds will be lower? And to what extent will they be lower?

So we do have these step-ups and step-downs, not on all the bonds. And some of the bonds are estimated about $50 million lower interest expense if we were upgraded.

50, 5-0?

That's right.

Thank you all for your attention and interest. As I mentioned in my opening remarks and from the discussion on this call, it's clear that we are heading on the right course. The industry backdrop has improved. The market trends are positive. ArcelorMittal is making progress, progress on our financial performance, progress on our Action 2020 plan, progress towards investment-grade balance sheet and progress in terms of our asset portfolio including Ilva. These are exciting times for the company and our shareholders, and I look forward to updating you as we move forward. With that, I will end the call by wishing you all a safe and enjoyable summer. Thank you very much.