Good afternoon, everybody. This is Daniel Fairclough from the ArcelorMittal Investor Relations team. This morning, we published our results for the first quarter 2020 alongside a Q&A document and a detailed presentation with speaker notes. The intention of today's call is not to go through that presentation again, but to move directly to your questions. As a result, we should be able to complete this call in about 45 minutes. If you'd like to join the queue to ask a question, please do press star one on your telephone keypad at any time. As in previous quarters, asking you guys to limit yourselves to one question has worked very well, so we intend to follow that format again this quarter. With that brief opening, I'll hand over to our Chairman and CEO, Mr. Mittal.
Thank you, Daniel. Good day and welcome everyone. I'm joined on today's call by Aditya Mittal, President and CFO, Simon Wandke, Head of Mining, Genuino, Head of Finance, and Daniel. Before we answer your questions, I would like to begin with a few remarks. It goes without saying that the recent focus for ArcelorMittal has been on addressing the impact of the COVID-19 crisis. First and foremost, addressing its impact on the health, safety, and wellbeing of our people. Then of course, mitigating its impact on our financial and operational performance. Our response to the pandemic has been swift and has been built on four pillars. The first is ensuring our operations continue to be safe and healthy working environments.
We are following government and World Health Organization recommendations and have implemented practical changes to our facilities, which fall into three distinct but equally important areas: enhanced cleaning and hygiene, social distancing, and personal protective equipment. The second pillar is to ensure our operations play a responsible and appropriate role in supporting their communities in response to COVID-19. I'm very proud of the actions our businesses have taken in this regard. Efforts have included a combination of financial and in-kind donations, including the donations of ventilators, test kits, face masks and shields, and steel for the construction of temporary healthcare facilities, and offering the use of our facilities to support local healthcare authorities. Our global R&D teams have also put their skills to excellent use. Designing and 3D printing ventilators and face shields, the latter of which we can produce 600,000 of a month.
The third pillar is aligning production with the reduced level of demand. This has meant idling assets across all our operating segments. The fourth pillar is to reduce all our costs, including the temporary reduction of fixed costs and suspending all non-essential CapEx. I believe you can see from our results today and our updated guidance, that we have made good progress in all these areas. I should also highlight that we entered this period from a position of significant financial strength. Our net debt is close to its lowest-ever level, and our liquidity position is strong. The environment in which we are currently operating is unprecedented. ArcelorMittal has faced several crises before, and we have all the management knowledge and experience required to navigate these exceptional economic conditions. Thank you. We are now happy to take questions.
Thanks, Mr. Mittal. Just to remind everybody, if you would like to ask a question, do press star one on your keypad. If I could reiterate the request that people limit themselves to one question in the first instance and then rejoin the queue. We have a queue of questions, and we'll take the first one from Seth at Exane, please.
Good afternoon. Thank you for taking our questions today. If I can start out with regards to cost savings, can you please give us a bit more color with regards to what efforts were made to cut fixed costs across the major regions, contributing to that guidance for a 25%, 30% Q-on-Q reduction in fixed costs? Obviously, the Q2 guide, it seems very positive. Can you then touch on whether or not there's risk of fixed costs under absorption re-emerging into Q3, assuming some of these government support measures likely fade? Thank you very much.
Sure. Thank you. First of all, I just want to open my remarks by thanking all of our employees for all their efforts that they have done, not only in making sure that all our employees are safe, but also in terms of navigating this crisis. I think one of the areas that you see that very clearly is what we have done in terms of fixed cost. I'll broadly characterize it into four. The first is labor. We have used the government-offered economic unemployment schemes, primarily in Europe as well as in North America, which has supported us in reducing these costs. Secondly, we have also reduced our repairs and maintenance in line with the levels of production that we're now witnessing in the second quarter.
Third, we also benefited from depreciation. In local currency, clearly this impacts the full fixed costs, you have seen weakness in currencies in CIS as well as in Brazil. Lastly, SG&A savings. Obviously, everyone is working from home, SG&A cost has reduced. Those four factors have variabilized, we like to call it internally, variabilized our fixed costs to the level of production that we are now incurring. In terms of region, it is actually quite similar across the region. I think everyone has done a similar effort in terms of variabilizing the fixed costs, the difference by region is not significant. The more significant difference by region is actually the level of production volumes. Clearly in Europe and in NAFTA, the reduction in production compared to Q1 is higher than what we see, for example, in the CIS business.
Thanks, Aditya.
Thank you.
We'll move to the next question, please, from [Alan] at ODDO.
Can you hear me? Yeah.
Yes, we can.
Okay.
Go ahead.
Thank you. Yeah, my question is really on, if you could come back on Ilva, EBITDA performance in Q1, and could you also discuss the new industrial plan and the implication of the expected cash out versus the previous plan?
Sure. In terms of Ilva Q1 performance, it's actually better than Q4. We have not provided you with a specific breakdown. Q4 was much more difficult for us, driven by the removal of immunity. Q1 has had better operating performance, and that has fed through in terms of results. Going forward, for rest of the year, we have been able to achieve the same actions that we have done across the rest of our business in terms of variabilizing our fixed cost base. We had the same impacts in Ilva into the second quarter as they have similar economic unemployment schemes. In terms of the overall agreement that we have reached with the government, we announced this after our year-end results. Fundamentally, we have a net purchase liability, which is in excess of EUR 1.1 billion.
The idea of the agreement is to convert that into equity. New shareholders would come in and convert our purchase liability into equity. The advantage of doing this is twofold. Number one, and I think most importantly, it's stakeholder alignment. I think for this facility to succeed in the medium to long run, we have realized that we need deeper and more engaged stakeholders. The second advantage is it provides us to lower our fixed cost base. We are working together with the government on a new industrial plan, a new business plan to make sure that the company is viable. If for whatever reason we do not arrive at such an agreement by November, then ArcelorMittal has a withdrawal right from Ilva.
Okay. Thank you.
Thanks, Aditya. We'll take the next question I have from Alain at Morgan Stanley.
Yes, sir, thank you for taking my question. My question is a bit of a high level question on the impact of the crisis on the European steel industry. Do you see any lasting implications on the industry structure? I would say more specifically on the capacity exit. Are any of these idled mills going to stay shut forever? What's your assessment there? Thank you.
Sure. Thank you for your question. I think like everyone, we'd be reluctant to predict how demand will shape up. I think there's limited visibility at this point in time. I do believe it's premature to talk about how demand will return and what impact it would have on facilities and operating utilization rates. As far as ArcelorMittal is concerned, I would point to a few things that we are trying to highlight. The first is we have provided guidance to you for the second quarter, primarily because we feel this was an uncertain environment and we wanted to give you a sense of how much effort we have put in to reduce our fixed cost base.
There are some indications that this may be the low point in terms of volume and activity level, primarily because you can see that, especially in Europe, the easing of lockdowns has begun. We see the same trend in some parts of the U.S. We also have automotive restarts occurring both in Europe as well as in North America. We have seen that in China, post easing of lockdown, demand levels have increased quite dramatically. Production levels have increased quite dramatically. One could argue that second quarter would be the low point of the cycle. Clearly this is a very uncertain and volatile environment, we all need to be prudent. Your question is more how does demand look like post that environment?
As I said earlier, it's premature to speculate, but clearly to the extent that we need to make variable cost savings or structural cost savings, ArcelorMittal would do that to adjust to the new demand levels.
Can you see any implication on capacity, long-term capacity in Europe specifically?
I cannot comment on others. To the extent that the demand level, I guess your hypothesis is that the demand level does not normalize to levels that we have seen in the recent past, clearly we would make structural fixed cost reductions to make sure that our businesses remain competitive.
Yeah. Thanks, Alain. We'll move now to the next question, please, from Rochus at Kepler.
Yes. Thanks for taking the question. Can you comment to what extent you expect, during the course of 2020, an appropriate step up of trade protection measures in the E.U.? I guess you commented earlier about potential anti-dumping investigation against Turkey, maybe you can elaborate a bit on that. How do you see, generally, your competitors or the market in Europe overall currently to adopt to the lower production levels, which you're seeing in the market? That's my question.
Hi, thanks for the question. I don't want to comment specifically on our competitors. We never do that. I can talk a little bit more about trade. You're right. The European Commission has launched an anti-dumping investigation on Turkey. As you know, they have to complete their study and then announce the findings. The other activity in the European Union is the review of the safeguard measures. As you know, the safeguard measures are a form of quota, to ensure that there's a level playing field, and we are not flooded with unfair imports. That is being reviewed as to how the level of safeguard can be reduced to match the new demand environment in Europe. Those discussions continue. You could have in Europe a tightening of the safeguard measures as well as further anti-dumping actions on different countries.
Okay. Thank you.
Thanks, Rochus. We'll take the next question, please, from Phil at KeyBanc.
Very much. Question is just on the question that Seth asked for just a little bit more color on the fixed versus variabilizing costs. The labor component you mentioned, so you mentioned labor MRO, local depreciation. How much of that fixed versus variabilizing you think is that labor component specifically out of the chunk, and that you were able to successfully take out this quarter?
I feel it's a very good question and we were trying to get a better sense of this. Fundamentally, I would say that labor is a large chunk of it, close to half of it, and the rest is R&M reduction that we have achieved due to SG&A savings, as well as the benefit of local currency depreciation.
Thank you.
Thanks, Phil. We'll move to Bastian, please, at Deutsche.
Good afternoon. I've got one question on the ACIS business, which was pretty weak in the first quarter. Could you maybe give us a bit of a sense how much of the contribution maybe came from the CIS operations in Ukraine and Kazakhstan, and how South Africa has been performing against that? I'm also maybe curious to get your view on what's currently going on in the Chinese market, where production rates seem to have actually been rebounding. Inventories are falling, yet the auto sector obviously is still running at a low utilization rate, albeit recovering. How do you see steel demand there at the moment? Where is that material going? So far, it's not really clear to me, what's been driving that rebound. Thank you.
Bastian, this is Genuino. I will address the first part of your question on ACIS. I think in Q1, some of the weakness that you see, are twofolds. One, of course, we highlighted in Q4 that we had some higher volumes coming from Ukraine, which would not repeat again in quarter one. That's one impact. Second, Temirtau was heavily impacted by the winter. We have seen very high levels of snow in January in particular. That has caused some production issues and as a result, we also lost some shipments there. Clearly South Africa was also impacted already by the lockdowns. On top of what was already a kind of a weak environment, we started to see already some impact of COVID-19.
Just to follow up on that very briefly. Is South Africa presumably losing money here, on EBITDA level? If so, maybe could you give us any sort of quantification?
As you know, South Africa is a listed company. They don't disclose their results now in quarter one. I'm afraid I cannot be very specific on South Africa at this point, Bastian.
Maybe you can answer the reverse and give us a number for the CIS business.
Sorry, I did not understand your follow-up question, Bastian.
Maybe you can give us the performance of the CIS business and how far they are maybe positive on a combined basis?
Clearly, CIS is positive on a combined basis. I think it's fair to assume that most of the contribution that you see there is coming from CIS.
Okay. Thank you.
Hi. In terms of your question on auto, in China, we have seen demand levels rebound in China, and it's not just auto. Clearly, we have seen the same in terms of industrial production. For example, we have seen the coal sector also consume similar levels of power. We've also seen crude steel production and crude steel demand also move up to much higher levels than what we saw in February and March and the month of April. Have I answered the question, or was there something more that you wanted me to address?
Yeah. Maybe you could give us your view. The reported inventory numbers seem to be coming down. I think, around February, it was pretty clear that there was quite a bit of inventory build, probably in the numbers we are maybe not directly seeing. I'm wondering, are you concerned that there is any sort of larger inventory overhang in the system in China?
You are right that there was a lot of inventory, in the beginning of this year, driven by COVID. The takedown of that inventory has been quite impressive and I don't have a crystal ball in front of me. I think these are all things which are difficult to predict. I think that the fact that not only is steel, but other sectors normalizing levels of demand and production should suggest that inventory levels should normalize.
Okay. Thanks, Aditya.
Sure. Bye.
Thanks, Bastian. We will move to the next question, please, from Alan at Jefferies.
Thanks. The indications for the steel shipments for Q2 were very helpful. Can you give us a bit of a sense within the quarter, what you're seeing in the order book and how April versus June is trending? Effectively, do you think the Q2 exit rate will be at a stronger figure than the beginning of the quarter?
That's a very good question. Yeah. To some degree that is true, but I would not suggest that's true necessarily globally, because also that would provide too much of guidance into the second half. What happened? Basically, European lockdown started in the second half of March. The U.S. followed a bit later, and Brazil followed a bit later, and the CIS obviously does not have the same level of impact. That is how it flowed through, in terms of our business. When I look at the shipment level of second quarter, clearly the flat business of both Europe as well as NAFTA is more impacted. If I look at the average of the two shipments are down more or less at the midpoint, including long at the midpoint of our guidance, which is down 30%.
If I were to look just at flat, that's down lower and obviously that is amplified by the automotive shutdown. Brazil levels are similar, but then CIS levels are muted compared to that average. Yes, there is a little bit of that inflection because you see European economies entering into slowly easing the lockdowns, clearly that is supporting demand. I think the biggest driver of that, which still has to be determined, is how strong is the automotive restart. I would say still early days. We provided a range of guidance, and I think that provides different scenarios. Generally, it's safe to assume that the endpoint is slightly stronger than the depths of the quarter.
It's very helpful. Thanks.
Great. Thanks, Alan. We'll move to the next question, please, from Luke at JPMorgan.
Afternoon, all. Thanks for the call. My question's on working capital. Clearly, a better than expected performance in Q1. Could you give a sense of how we should be thinking about the EUR 1 billion release this year, just in the context of this being given, or first articulated pre-COVID-19? Clearly there will be additional upside based on lower volumes. Also maybe just any granularity you can give on the levers that can be pulled, just because inventories look like they dropped fairly significantly already in Q1, and receivables are already at a low level. Thank you.
Sure. Yeah, in terms of working capital, you are right, we have not updated our guidance. Our guidance at the beginning of the year is based basically on the efficiency of working capital. That still holds true. We believe that we can be more efficient users of working capital, and then therefore, we expect a EUR 1 billion release. In terms of the market effects, I think clearly, it is very early days to predict volume and price levels. As you know, the last 90 days, i.e., the fourth quarter, is a prime determinant of how working capital will shape up. Specifically in terms of Q1, I think you are right. I think the organization did a tremendous job in ensuring that as COVID-19 began, we were not producing extra, and we actually had the ability to manage our inventories.
As a result, you don't see the working capital build into Q1, which is seasonal norm. In terms of the second quarter, we're not providing quarterly guidance, but we also reduced our purchases quite a lot. There is also the decrease of payables, which one must keep in mind as well. Working capital, the story is that we still expect, due to efficiency, to release cash in 2020, but difficult at this stage to provide you with specific guidance.
Thank you.
Great, thank you. We'll move to the next question, please, from Carsten at Credit Suisse.
Thank you very much. My question is on the portfolio optimization program. That seems to be still on track. At least you keep it and will be finalized until mid-2021. My question to that one is it the right environment right now to sell assets? How do I have to interpret the mid-2021 guidance for finishing it? Could be some of the projects pushed out to 2021. How much do we actually want to achieve in 2020 out of the portfolio optimization program? Thank you very much.
Sure. We launched the program, as you know, first in May 2020. It's a two-year plan. We've done about EUR 600 million. We have about EUR 1.5 billion to go. We have multiple options in this. It's not like we're only focused on one asset or one opportunity. So far, the interest of buyers is still there post-COVID-19. So far, we still find that the values are attractive to ArcelorMittal, and I do appreciate your point. We are not in a fire sale mode, and I don't expect a fire sale on any of these assets. I do believe that we can get fair value and that we can still achieve and implement this program by May 2021.
Okay. Thank you.
Thanks, Carsten. We will move to the next question, please, from Myles at UBS.
Great. Thank you. I was just wanting to understand the curtailments that you're making, obviously cutting production in the second quarter by about 30%. What proportion are hot idled versus cold idled? How quickly will it be to ramp the blast furnaces back up? Just related to that as well, should we assume that you manage your supply in line with market demand, or because it looks at the moment like you're doing more than your fair share of curtailments? Thank you.
Yeah. In terms of banking furnaces or idling furnaces, to be honest, it depends by region, and what our technologists want to do. It's not uniform on a global basis. For example, in North America, there's much more banking than in Europe, as an example. Nevertheless, the ability to restart a furnace and restart production, it's still relatively short, right? Even if you idle a furnace, because we're hot idling furnaces, we're not cold idling furnaces. In terms of our response, what we have done is we have reduced our production levels in line with demand levels. We do not believe that we have a deterioration in our demand level, i.e., we do not believe that we should be disproportionately losing market share as a result.
I think on the contrary, the fact that we have reduced our cost adjusted to the crisis should allow us to maintain the same market presence. Perhaps there are changes because we do have, or differences because we do have automotive exposure, quite significant amounts of automotive exposure in Europe and in North America. Apart from mix issues, I would not expect any change from the rest of the industry or the countries in which we operate.
Okay. Thank you. Just to be clear on the cold idling, have you done any cold idling at all so far? How long can you hot idle before you'd have to cold idle?
Once you cold idle, you have more or less, if we're talking the same terminology-
Yeah
-once you cold idle, you're basically taking down the furnace. If you want to bring it back, you have to realign the furnace, i.e., put the bricks back in, then heat it up, then restart, and that can be a 3-9 month job, depending on what is the state of the furnace, are the bricks available, et cetera. Hot idle is when you keep the furnace warm, you would have to, obviously, there's a slow restart, there's a restart, then you begin production again. It's not a matter of time. You can keep furnaces hot idled for a considerable amount of time. It's really, what is your market outlook? If you do not want the furnace on a medium-term basis or even longer-term basis, that's when you move from hot to cold idling.
Okay. Thank you.
Thanks. Thanks, Myles. We'll take the next question, please, from Grant at Bloomberg Intelligence. Are you there, Grant?
Sorry, we do not have Grant in the question list anymore. Grant, if you would like to ask a question, please press star one.
Okay. I think we can move then to Christian at SocGen.
Yes, thank you. I wanted to ask you about your Canadian iron operations. Are you getting any production interference in Canada because of the pandemic at all? In term of your end markets, do you foresee some reduced demand levels in the second quarter, which may have an impact on your performance? Thank you.
Christian, it's Simon speaking. Let's talk about Canada. Essentially in Q1, the performance has been impacted by COVID right at the end of the quarter. I'll just go through a couple of dates here. You recall the government of Quebec introduced an initial rule to reduce intensity of workforce on the ground around the 24th of March. At that point, we had contractors on site. We had FIFOs as usual, and we were asked to stand down all those areas, and reduce our workforce. Of course, we operated at a lower level for a period and then stepped back up when the government asked us to, as with the mining industry, to rekindle and get going safely from about the 15th of April on a steady basis, which we complied with. We're now running full.
Whilst demand has changed within the group for product in which we produce, and whilst we've lost some production, we're running full on a confident basis. We see this market strong in China. China's coming back. I think there was a few comments a little earlier. You're also looking at iron ore stocks at the ports, quite low at the moment, around 115 million tons. Seems as though their value chain has been pulled through for the mills. Blast furnaces are running at high capacity, relatively higher than the eighties it seems. We see a relatively easy transition to move product that may not be required in Europe in Q2 currently to move to other markets, particularly China.
Thank you.
Christian, did that sort of nailed it for you?
Yeah, absolutely. Thank you very much.
Great. Thank you. We will move to the next question, please, from Olivia at Bank of America.
Hello?
Hi, Olivia. Go ahead.
Hi. Good afternoon. Thanks for taking my question. I have a question regarding the variableization of fixed cost. I think for Q2, probably an important driver behind this is the state funding from the government to partially subsidize the labor cost. Going beyond first half, particularly in, I guess, U.S. and ACIS, are you still able to get any sort of state support? Basically, how long do you think you can still variableize the cost even if COVID-19 situation and all these lockdowns last into Q3 or Q4? Thanks.
Sure. Thank you. Thank you, Olivia. First of all, I think we're very grateful for governments who have stepped in and provided economic unemployment support. I think that has been important not only for our company, but for our sectors and the economy in general, to buffer the impact of the virus on the economy. In terms of what we have done, as you rightly pointed out, we have utilized these programs in Europe as well as in NAFTA, and to a lower degree in other markets. I don't believe we have done anything in terms of the CIS. I don't believe there is economic support for unemployment in the CIS, but we've still reduced our fixed costs there by using R&M, SG&A, as well as the fact that those currencies have also weakened.
In terms of going forward, based on the discussions we have had, we see that this thing has legs, or at least has the same amount of duration as a weakness in demand. I don't expect that this economic unemployment support would be removed before demand levels normalize. From our standpoint, again, there's always limited visibility and no one can predict the future. I do not see that much of risk that these programs end before we can normalize our production levels.
Yeah, that's clear. Thanks very much.
Sure.
Great. Thanks, Olivia. Can we take the next question, please, from Seth at Exane?
Sorry, my follow-up question was just answered. Thank you very much.
Thank you, Seth. Okay, so we do have some more follow-up questions. I think we'll move. Next up is Carsten at Credit Suisse.
Thank you very much. One follow-up question on India, because your volume guidance was pretty much on your businesses, I believe Europe, Americas, and CIS, but we haven't really covered India yet. Could you give some more detail how you expect shipments to evolve in the Essar joint venture, in the second quarter, and if possible, post the second quarter? Thank you very much.
Yeah, Carsten, that's an excellent question. Just for the benefit of everyone, even though it's in our ER presentation, I'll just take a few minutes to update everyone on India. That's the latest acquisition. We completed it in December, and we have had more than a quarter now operating the business, but I'll talk about Q1. The business did about $160 million of EBITDA, which is run rating at about $540 million. In terms of the cash requirements of the business, it's less than $250. You can see that there's a strong EBITDA to cash conversion ratio. We also finalized our long-term financing for that with a Japanese bank, so that helps that the cash requirements are low.
This includes maintenance, CapEx, interest, tax, the usual, to get to what is the free cash of the business. We also made strategic progress in the business. We bought 5 million ton mine for less than EUR 15 million. We have also acquired a relatively new gas power plant for 500 MW for about EUR 60 million. As we look forward, we see that the cost position of our business continues to improve. We see the cost of iron ore in India has come down, because a lot of the iron ore in India, as you know, is stranded. On top of it, we have seen the price deck for natural gas come down, that has made our DRI facilities that much more competitive. Thirdly, we have the advantage of the fact that the facilities that we have purchased are coastal.
Hazira, the steel making, is on the coast, as well as our pelletizer facilities. In the second quarter, we have exported a lot of pellets. We have also exported steels to nearby markets, Southeast Asia and others. We have not provided a specific guidance for India as it's not consolidated. I think at every quarter we can update you on its performance. In terms of India as an economy, clearly it's a developing economy. There is pent-up demand. I believe post-COVID, the government will be focused on restarting infrastructure spend and restarting consumer spend, and trying to get back to a more normal environment as well.
Okay. Thank you.
Thanks, Carsten. We will move to another follow-up from Alan at Jefferies.
Thanks. Just on Ilva, quite a positive step change in the safety performance there. Is there anything within those ratios that's a function of lower activity, or do you think you've achieved some breakthrough in perhaps implementing some best practices in that facility?
In terms of your question, you're referring to Q4 versus Q1 performance?
Well, Q1, really compared to any of the quarters from all of 2019.
No, that's a fair comment. I think what we have achieved in Q1 is higher levels of production. Clearly, we began to implement our improvement program, which, as you know, started in the first half of 2019. We were hit with two events. The first was we had a hurricane which impacted our raw material station, our raw material unloading pier. As a result, we were producing less in the third quarter. There was a lot of political noise around immunity, which we finally lost in November. As a result of those factors, we did not have very normal operations in the second half of last year. Now entering into Q1, those issues are largely behind us, and so we had more normal operating environments. On top of it, some of the savings that we had begun to implement kicked in.
I think that's what you are seeing, but I don't want you to get carried away. Ilva is still loss-making in Q1.
Sorry, I was actually referring to the safety ratio. Is there any best practices there? That's been quite a big change in Q1.
Yeah, I think it's a combination of everything I have said. I think clearly when these exogenous circumstances were behind us, we could focus on managing the business on a day-to-day basis with more rigor, and I believe you see that in the safety numbers as well.
Yeah. Thank you.
Sure.
Perfect. Thanks, Alan. We'll come back to Grant at Bloomberg Intelligence. Hopefully you're there now, Grant.
Hi. Good afternoon, everyone. I hope you can hear me.
We can. Thanks.
Great. Thank you. Thanks for taking my question. Regarding the Action 2020 program, which I'm just trying to square in my own mind, you've obviously got the variabilization of fixed costs, which is very understandable in this environment. How should I think about the targets under the more permanent, let's say, reduction of costs? I'm just trying to square what's temporary and what's permanent going forward. Thanks very much.
Thank you. Thank you for joining us today, and thank you for your question. At this point in time, all the fixed cost savings that we have outlined are temporary in nature. Clearly, it's very hard to predict the future. We all have limited visibility. To the extent that demand levels don't come back to pre-crisis levels, we would review how to make some of these temporary fixed cost savings more permanent in nature and look for structural cost measures.
Okay. Thank you. Can I just clarify in my own mind and maybe I've missed something, is all the targets that you previously outlined under that program, have you formally removed those?
That's our Action 2020 target, and two-thirds of that is variable, and a third is fixed cost. I think the focus of the whole organization has been to adjust our cost base to the new operating environment in the second quarter, and we talked about how we've done that and I think the teams around the world have done a good job. In terms of the variable cost piece of Action 2020, that's more difficult to achieve when you are not operating at high utilization rates. Yes, some savings can be made, but it comes through when operating rates are higher. I would expect that we would achieve the variable piece more in 2021 than in 2020 or the latter half of this year, depending on when operating rates normalize.
The fixed cost portion, yes, we should still continue to make progress on that this year in terms of Action 2020.
Got it. Thank you very much for the explanation.
Yeah, sure.
Thanks, Grant. We'll move to a follow-up from Rochus at Kepler.
Thanks for taking the follow-up. I want to come back to the previous question on Chinese inventory. I guess we all see this data in China on the steel trader and mill side, where inventories are ticking down from record high levels. What do you observe from your market knowledge in terms of the direction of the inventory which are stuck at so-called unmonitored locations? Are you seeing a similar trend in terms of inventory decline? Considering that total inventory have been 2 times higher than normal at this time of the year, what kind of risk do you see that this will end up sooner or later in the world market when world ex China opens up again? Would you expect more production discipline in China to absorb that?
Yeah. It's very hard for us to see through that inventory as well. I do not have any more information than what you have. I think we have seen the drawdown. We have seen that IPS picked up. We have seen coal production pick up. We have seen also automotive producers reporting that demand for automotive is also normalizing. Perhaps Daniel has more insights, so I'm just going to turn it over to Daniel.
Thanks, Aditya. No, I think we're all watching the same data. We obviously saw the concerning accumulation of inventory during the crisis, obviously. Maybe unlike some of the other regions, production was maintained despite the drop-off in demand. There was quite a sizable accumulation of inventory as we went through January, February. The pace of the drawdown, as you identify, has been very rapid, both at the service centers, but also even more dramatically at the mills over the past seven weeks. If that rate of drawdown continues for the next four, five, six weeks, then you'd be back in that sort of normal range of inventories in China. That's something that we're watching, and we're obviously also watching the export pattern from China. That was impacted during the crisis when they struggled to get material to the coast.
We have seen a little bit of an increase the past couple of months, but no big concerning increase in those exports. That is something that we'll be very watchful of in the coming periods.
Okay, thanks a lot.
Great. I think we've got one final question to come from Myles at UBS.
Great. Thank you. There's actually two questions, but they're not particularly quick to answer, I'm sure. Just on the CapEx side, first of all, $2.4 billion, it's pretty low. How long can you keep CapEx at this sort of level? Is this a one-year wonder, and it's going to normalize back to 3.5 next year? Or can you keep it at this level for two, three years if we have more of a depression-type outlook? Secondly, just on the new credit facility, I don't quite get why you went for a 12-month facility. Are there restrictions around the $5.5 billion facility, means you don't want to draw it? Is it to protect the credit rating? What was the rationale with that new bank facility?
Okay. Thank you for your question. In terms of CapEx, there is some growth CapEx in the $2.4 billion number as well. We're completing the Mexican hot strip mill. We are more than halfway through that. There's not that much CapEx left, and we are continuing with the agreed projects in Italy, as well as some of the CO2 projects that we have started. In terms of how long can we maintain this, I think we can maintain this for some time. I don't want to come out with a very specific prediction of how long, but it's not that we are not maintaining or doing essential CapEx. We are doing that. I do expect that assuming production levels have not normalized, we could maintain this CapEx. Assuming production levels normalize, clearly, I would expect and imagine that the maintenance CapEx levels would increase.
Also, I would just add that if the economic environment remains weak, then you do have two effects on our CapEx line. Number one, you can negotiate better prices or costs from CapEx suppliers. We do get some benefit from the devaluation that has occurred in some of the markets in which we operate. Coming back to the liquidity line, no, there's no restrictions on the $5.5 billion liquidity line we had. We just thought we would bolster our liquidity, that we further strengthen our balance sheet. We do have a commercial paper program, which saw some weakness, I think that has also normalized. Just entering into the crisis, we thought it's a good idea to enter into a liquidity line. This was signed obviously, this week, obviously the negotiations began five, six weeks ago, right?
It was at that point in time, we thought one of the actions we can take is to strengthen the liquidity line. It is not at a significant cost. None of our liquidity lines are drawn. It is just there as a backup facility. Perhaps it helps on the margin for the rating, but it is not driven due to rating agency considerations.
Okay. It's linked to the commercial paper more than anything else, which makes sense completely. Thank you.
Yeah. At that point in time when we began the discussions, we thought it was appropriate to improve our liquidity, as we'd entered the crisis. Maybe today you look at it and say, okay, the situation globally is, in terms of CPs and other lines, are not so bad. Perhaps the requirement is not as great. I think it's very good that we have it. It ensures that we have strong liquidity through this period.
Excellent. Thank you.
Sure.
Thanks, Myles. There is one final question actually, and it's a follow-up from Phil at KeyBanc.
Hey, thanks for taking my follow-ups, and I'm all set. Appreciate it.
Okay. Perfect then, Phil. I'll hand back to Mr. Mittal for any closing comments.
Well, thank you everyone for your continued interest in ArcelorMittal. I wish you and your families all the very best in these difficult times. Keep well. We will speak soon. Stay safe.