Century Aluminum Co (CENX) 2018 Q1 法說會逐字稿

Welcome to the Century Aluminum Company First Quarter 2018 Earnings Conference. (Operator Instructions)

As a reminder, this conference is being recorded.

I would now like to turn the conference over to our host, Finance Manager for Century Aluminum, Mr. Peter Trpkovski. Please go ahead, sir.

Thank you very much, operator. Good afternoon, everyone, and welcome to the conference call. I'm joined here today by Mike Bless, Century's President and Chief Executive Officer; and Shelly Harrison, Senior Vice President of Finance and our Treasurer. After our prepared comments, we'll take your questions.

As a reminder, today's presentation is available on our website at www.centuryaluminum.com. We use our website as a means of disclosing material information about the company and for complying with Regulation FD.

If you take a look at Slide 2, please take a moment to review the cautionary statement shown here with respect to forward-looking statements and non-GAAP financial measures contained in today's discussion.

With that, I'll hand the call to Mike.

Thanks very much, Pete, and thanks to all of you for joining us this afternoon, as always.

If we could turn to Slide 4, please. I'll give you a quick rundown the last couple of months. They've obviously -- they've been busy ones. And most importantly, as I'll note in a couple of moments, we had a very good quarter in the operations. Safety performance was acceptable, and we had a generally stable process and good efficiencies across the department at each of the plants.

Financial results for the quarter came in just as we had expected. As we had forecast, realized higher metal prices were significantly more than offset by raw material price increases. Carbon costs were up as expected, and the same is true for the higher alumina costs on material purchased back when the market was high, as you'll recall, in the late months of 2017.

This will actually go the other way in the second quarter results, as the normalized prices that we saw at the beginning of the year roll through our P&L. Of course, the market's moved meaningfully since then, and I'll comment on that in just a couple of minutes.

Against that higher alumina price, we captured only a portion of the higher metal price in the first quarter, and that's, of course, due to the fact that most of our sales contracts, in fact, virtually all of our sales contracts, are priced on 2-month lag. We'll see close to that full amount in the second quarter. And Shelly, in just a couple of minutes, will provide you all the detail on the various price movements, both from Q4 to Q1 that we just -- that we're reporting today, and then she'll also give you some estimates on those same movements from Q1 to Q2.

The alumina price did develop in -- late in 2017 and early into this year, precisely as we predicted. The index had come down from just shy of $500 a metric ton to about $350. And based on a couple of transactions that yet had -- hadn't been incorporated into the index, we think it still had a bit further to go, and this, we believe, was a rational reference at the time of the alumina to the metal price.

Of course, there was an unexpected development in early March that sent the price quickly back to the prior levels in the high $400s. That, of course, through the Alunorte refinery in Brazil, which was forced by authorities to curtail 50% of its production. This was due to a 100-plus year rain event and concern by the authorities relating to untreated wastewater discharge. You've obviously read extensively about this.

Alunorte, as you know, is the largest refinery in the world, and the sudden removal of 3 million metric tons in the Western world traded alumina market, of course, had a very significant impact.

The majority owner of this excellent refinery has studied the situation and has said they believe the conditions are now safe to restart. We understand the discussions are ongoing, and we firmly believe that logic dictates it's merely a matter of time before it restarts. The refinery is simply too important to the local economy.

The industry was thrown into further uncertainty, of course, with the implementation of sanctions on various Russian entities and individuals several weeks ago. And more recently, the deferral of the effective implementation date and a potential pass for exemptions has caused the market to adjust in the other direction. So sitting here today, metal's currently trading just above $2,300 a tonne. The alumina price has posted at about $640, but the forward alumina price is down to $500 by the end of the quarter. Even that forward price represents over 20% of the aluminum price, and that's far above the level of the -- of a rational market.

We continue to believe, as do most industry participants, that the right value for alumina is in the range of 16% to 17% of the LME price. That would indicate mid-$300 to the current metal price. It goes without saying, we're going to see price volatility based on actual developments and rumors here over the coming weeks and months.

Century is well supplied for the coming months, and we're working on longer-term plans should the current situation persist. We'll obviously feel the financial impact as we do buy every day, as you know, with reference to the index price. But we do believe this situation will be short lived.

We obviously reached another major milestone in the industry when the Section 232 tariffs became effective on the 23rd of March. All primary aluminum imports today are subject to the tariffs other than production in countries that have temporary exemptions. Thus far, the tariffs have had -- an expect -- the expected impact on the U.S. market with the immediate announcement of production restarts, and I'll comment in just a couple of minutes on our own plants.

The structuring of any exemptions to the tariff will obviously be critical to ensuring the continued realization of the administration's desired outcomes, and thus far, we're seeing exactly that.

A couple of days ago, on Tuesday, the government announced that all exempted countries, other than Canada, Mexico, and the European Union had reached agreements in principle on quotas. Argentina later announced their quota will be equal to their 3-year historical average level of imports. And from what we understand, any further exemptions for the remaining countries will be based on that same concept, that is limiting imports to historical levels or below. Exemptions for the identified countries structured this way will continue to backstop the administration's goal of supporting U.S. production restarts and, importantly, the long-term competitiveness of the industry in this country. The administration's clearly on record they will ensure that all primary aluminum imports will be subject to either tariffs or quotas. And we're very confident no action that would dilute the tariffs' intended objection -- objective will be taken.

Bottom line, we're confident that we'll soon get through these 2 near-term issues: number one, the finalization of the tariff regime; and number two, the clearing of disruptions in the alumina market. And for those reasons, we're proceeding apace with our plan to restart the 3 curtailed potlines at Hawesville.

The restart activities are proceeding on budget and ahead of the scheduled plan. The first line will be restarted before July and will reach its full 50,000 metric ton annualized capacity during the third quarter.

We've also advanced a high confidence plan to restart the last 2 curtailed potlines on an accelerated time schedule. We told you last month the plan should achieve full capacity by the back half of 2019, and we now believe we'll get there reasonably well ahead of that schedule. We're also very close to making a decision to invest in the new cell technology that we've told you about. The results of the 5 R&D cells continue to far exceed the model's expectations.

We're currently finalizing all these plans, and we'll provide you an update when we release our second quarter results. And that will include, of course, the timing for the restart of the last 2 potlines, when the incremental volume will come in and, of course, the timing of the spending. And for now, Shelly will give you some data on what we expect the expenditure during the second quarter will be.

It goes without saying that we're proceeding with this program during some uncertain times. But as I said, we're confident in the resolution of the tariffs in the alumina situation. It's serious incremental value to our share owners to get this production online as quickly as practical. And as a reminder, this program is quite flexible. We can moderate or even stop it literally on a day's notice.

Lastly, we continue to search for a new power contract to enable the restart of the curtailed potline at Mt. Holly. So let me remind you about the proposal that we've made. We've told you about this last time. We've made a proposal that would have us purchasing 100% of the power requirements for the entire plant from the competitive wholesale market. That, of course, would enable us to restart the curtailed potline.

We pay the local power company the same unit transmission fee but obviously on significantly more power. And we make a small additional payment that would be required to get the power company's revenues from us equal to what they're receiving today under the current agreement.

As you know, the local power company has long said it's not able to let us import more power, as they have contended the use of the incremental transmission capacity would hurt their other customers. They've recently publicly testified that they've never sought to calculate how other customers may or may not be hurt. We've long maintained the data show that there'd be no harm whatsoever. They have thus admitted they have absolutely no basis for refusing our proposal. As you'd expect, we're a bit perplexed by this development, but we do remain confident that logic will ultimately prevail.

And with that, I'll turn it over to Pete to give you some data on the industry.

Thanks, Mike. If we can move to Slide 5, please. I'll take you through the current state of the global aluminum market.

The cash LME price averaged $2,159 per tonne in Q1, which reflects a 3% increase over Q4. The LME price on a 2-month lag basis was up quarter-over-quarter 2% and averaged $2,129 per tonne. As Mike discussed, there has been a lot of news driving significant volatility in our markets over the past couple of months. As a result, aluminum prices have seen over a $600 per tonne range just in the month of April, averaged $2,250 per tonne for the month and are currently sitting above that level.

In the first quarter, regional premiums averaged approximately $0.144 per pound in the U.S. and $168 per tonne in Europe. However, spot premiums are significantly up and are currently approximately $0.22 per pound in the U.S. and $240 per tonne in Europe.

In the first quarter of 2018, global aluminum demand grew at a rate of 4% as compared to the year-ago quarter. We saw about 5% year-over-year demand growth from China, about 3% growth in Europe and around 3% growth in North America as well.

Global production growth was flat in Q1 versus the same period last year. This was driven by the winter heating season in

(technical difficulty)

Capacity cuts in China. However, despite these actions, China still added a net 3.8 million metric tons of smelting capacity during [2017].

On March 23, the U.S. implemented a 10% tariff on all primary aluminum imports into the United States in order to stop the flood of foreign metal that has been destroying the U.S. aluminum industry and threatening our nation's national security. These tariffs are now in place and working as intended.

The U.S. government has issued temporary exemptions from the tariffs to Canada, Mexico, the European Union, Australia, Argentina and Brazil while it negotiates quotas with these allies to restrain imports, prevent transshipment and protect the national security. It announced on Tuesday that it has already reached agreements in principle on quotas with Australia, Argentina and Brazil and that it had agreed with South Korea that its aluminum imports would be subject to the tariffs in full. It is clear that the U.S. administration understands the importance of having an effective tariff structure, and the administration officials have reiterated that any permanent exemptions will be subject to quotas to ensure the tariff regime remains effective in protecting U.S. national security and causing U.S. production to restart.

With that, I'll turn it back to Mike.

Okay, Pete. Thanks very much. If we can turn to Slide 6, please. Just a couple of comments on the operations, and then I'll turn it over to Shelly to go through the quarter.

As I said, we're pleased with the performance in the operations during the quarter. And let me just go give a couple of comments. Safety there, you see a slight downturn at 2 of the plants. Obviously disappointed to see that. I would note there that, that's simply a reflection of 1 incremental incident at each of those plants Q1 over Q4.

More importantly, we're making great progress toward the continuous and long-term improvement of the safety environment across the company. A couple of examples. One at Grundartangi. They're currently going through a multiyear reinvigoration of an already very fine safety culture and processes. At Hawesville, we've gotten appropriate and significant focus on the safety environment during the pot restart process there. You've got a really complex environment in that plant with a continuing operation existing side by side with the complex restart project.

As you can see, production was good across most plants. Hawesville lost a couple of cells in January during, as you recall, a bad snap of very cold weather. But those cells came very quickly back into service, and so you won't see that again in Q2. Production metrics, excellent, stable and favorable across the board, as I said earlier.

Moving down to controllable cost performance. Generally, it was good across the plants. Of course, the major mover during the quarter, as we expected, was raw materials, and we also had an impact of the cold weather on power prices in January. As you recall, we talked about each of these factors and the expected impact on the first quarter results when we released fourth quarter results in February. And Shelly, in a minute, will provide detail on the financial impact, in fact, during the first quarter.

Let me just make a few comments at the plant level on controllable costs. Remember, this chart shows conversion costs, so alumina has always been excluded. Again, Shelly will obviously comment on alumina costs.

As you can see, good performance at the plants in Kentucky. The issue at both Mt. Holly and Grundartangi was labor costs. There's really no factor as we've dived through the numbers at Mt. Holly, and no worrying trend has been detected as of now. But of course, we're watching it closely.

At Grundartangi, we are seeing a trend of higher labor costs. It's got 10% higher per metric ton of aluminum production labor costs Q1 over Q4. That's embedded in that number there. It's the result of a very hot economy in Iceland, I'm sure as you've read, most of you or maybe traveled, mostly fueled by tourism. Wage inflation in the local currency has been up between 6% and 8% in each of the last couple of years. Add to that a strengthening of the Icelandic króna by over 25% over the last couple of years. That produces a real issue that we're working through. We need to address that issue in order to preserve the long-term competitiveness of this excellent plant. We're confident we'll be able to do that.

With that, I'll give you to Shelly.

Thanks, Mike. Let's turn to Slide 7. I'll take you through the high-level results for the first quarter.

On a consolidated basis, global shipments were essentially flat quarter-over-quarter. But realized prices were up 5%, reflecting higher lagged LME prices and premiums as well as some improvements in product mix.

Looking at operating results. Adjusted EBITDA was $22 million this quarter, and we had an adjusted net loss of $0.04 per share. In Q1, our only adjusting item related to a lower cost to market inventory adjustment, which was a noncash benefit of $3 million in our reported results.

Turning to liquidity. Our cash balance decreased to $131 million as a result of a significant working capital build in Q1. The working capital investment was primarily driven by an increase in direct sales to end-use customers with longer payment terms.

Okay, let's go to Slide 8, and I can walk you through our Q-to-Q bridge of adjusted EBITDA. During Q1, we generated $22 million of EBITDA as compared to $60 million in Q4. As expected, the $38 million decrease was driven by $47 million in higher raw material prices as well as $6 million in higher U.S. power costs, primarily as a result of that cold snap that we talked about on our last call. These raw material and power price increases were partially offset by an $18 million benefit from higher LME prices and premiums. The $6 million increase in other operating costs was primarily driven by higher labor costs, of which over half related to Grundartangi, as Mike discussed.

Alumina costs for Q1 were based on a realized undelivered price of $435 per tonne, which was in line with the 3-month lagged index price of $445 a tonne. As expected, this was up significantly from the Q4 realized price of $338 a tonne. For Q2, we expect our realized alumina price to decrease to $382. This decrease of roughly $50 a tonne from Q1 should improve Q2 EBITDA by about $15 million.

In addition, LME prices and regional premiums have increased meaningfully over the past several months. Since our sales contracts average a 2-month lag in pricing, the relevant period for our Q2 results is February through April. For this period, the LME price was up about $40 a tonne. The Midwest premium was up $0.08 a pound. That's about $175 on a per-tonne basis. And the European Duty Paid Premium was up $20 a tonne. This is all versus the comparable period for Q1. As a result of the LME and regional premium improvement, we expect to see an EBITDA benefit of about $25 million next quarter.

So in total, we expect the Q2 EBITDA benefit from these changes in selling prices and realized alumina costs to be around $40 million.

As Mike mentioned, we anticipate that the first cells from the Hawesville restart will come online towards the end of the quarter, and we expect the impact on Q2 EBITDA to be a slight negative, as some of the training and labor costs will precede the benefit from the additional volume.

Okay, let's turn to Slide 9. We'll take a quick look at cash flow. We started the quarter at $167 million in cash and ended March with $131 million. As I mentioned earlier, we made a significant investment in working capital during the first quarter, as our 2018 sales contracts include higher volumes sold directly to end-use customers with longer payment terms.

On our last earnings call, we anticipated some reduction to inventory working capital in Q1, as alumina prices had fallen quite a bit from year-end levels. But the reduction at Alunorte that Mike mentioned caused alumina prices to rise, which drove our inventory balances back up.

In addition to the working capital investment, we also spent $3 million in CapEx during the first quarter. For Q2, we expect to have cash spending of roughly $20 million [that'll be capitalized] related to the Hawesville restart.

And with that, I'll hand it back over to Mike.

Thanks very much, Shelly. We appreciate again everybody's attention today. And now we'd look forward to taking your questions.

(Operator Instructions) And we'll first go to the line of Novid Rassouli with Cowen and Company.

Mike, so you mentioned that you don't believe the current environment for alumina prices will persist. I just wanted to see, we've had some developments on that front, given the Norsk Hydro call. I just wanted to see, what are your expectations for when you think is most reasonable for prices to trend back to the lower levels before all of this started?

Wow, that's a great question. I wish, without ducking it, I wish I had a better sense of the answer. The timing is going to be all -- all I can point to, Novid, is a couple of things. One is, as you know, it's fallen from its high. So it peaked above $700, and it came down a good chunk. I think, guys, it was sometime last week at this point, so maybe 5 or 6 or 7 trading days ago, posting days ago, I suppose, one should say, for the index. And that was -- at least market participants, Novid, believed that was due to the more accommodative, perhaps, is the right word -- words coming out of both the individual involved. Obviously, the majority owner, ultimately, (inaudible) and the U.S. government relating to the sanctions.

We think the real bottom will fall out of it when there's positive developments on Alunorte because, if you remember, the prices -- as I said in my comments, the price had, as we expected, come down to where we thought it ought to have been bid in the mid-$300s. And it was kind of sitting there for some period of time, a couple of months, and then it rocketed right up to, if I recall, just shy of $500, like $480, $490, guys? Yes, $480, right on the Alunorte development. It's -- we were talking about this the other day. It's kind of difficult to believe that, that announcement on Alunorte was only 8 weeks ago. It's -- so it's reasonably fresh.

And so we think, other than sanctions, of course, we've had 2 issues, as we said, as I said. And other than Alunorte, the price goes back into the mid-$300s. So I guess, that's what we would be looking for, look for some positive developments on Alunorte. You saw what that did to the price. It was up $150, give or take. And then the rest of the ride up was a reaction to the sanctions. I wish I could answer your question better. I would say, fundamentally, we remain convinced that the price goes back at a $2,200, $2,300 LME environment, as we've been seeing for the last couple of weeks, goes back into the mid-$300s where it belongs.

Right. And you generally realize kind of your alumina prices on a 3-month lag. I just wanted to see if, given this spike and the dislocations that we're seeing, extreme volatility, have -- is there any reason to believe that there would be a change in lags and how that pricing will flow through your P&L and how you'll realize that?

No, that's a really good question.

Yes, it can change a little bit from quarter-to-quarter just based on shipments and inventory levels. But on average, 3 months is still going to be a good reference point.

Yes, I guess, just speaking from an operational standpoint, it's an excellent question. I'll cut it into 2: operational and pricing. From an operational standpoint, all else being equal, of course, with a nod to liquidity and work in balance sheet, we would probably prefer -- we would prefer, in these times where supply lines are tight, if we can pick up a cargo or 2 that we otherwise wouldn't have had on hand at any of our plants, we'd probably do that, as I said. Looking months and months and months and months and months and months ahead, we're fine from a physical supply standpoint.

So could a couple of days more inventory -- of alumina inventory on hand cause a distortion in that 3 months? I think, as Shelly said, it would be marginal. Otherwise, as we're looking to replace cargoes, we, as other people, are looking at Chinese material, for example, that trades at a discount to the posted price that you see every day. And so that's the only other factor there. Again, whether at the margin that makes much of a difference, I couldn't say at this point in time. So that's a long-winded way of saying probably not.

Sure. And the $47 million, what was the portion of that, that was alumina?

Of the movement quarter-to-quarter?

Yes, of the bridge, yes.

[I think $30 million, alumina; $17 million, carbon does that sound right?]

I think it was closer to $40 million on alumina.

$40 million. Here we go. $35 million and $12 million. $35 million, alumina; $12 million, carbon.

Perfect. Okay. And Mike, I just want to make sure, I think I heard in your initial comments you said -- you mentioned something about taking alternative measures if prices remain elevated. I just want to see, was that about alumina? Did I hear that correctly? And what measures would those be?

I'm not sure what's -- was it, Novid, specifically related to the Hawesville restart program?

It might have been. I might have misheard you, though. But as far as just alumina procurement, as far as the restructure at Hawesville, nothing is at risk of not being able to actually procure for all those restarts?

Correct. Absolutely correct.

We'll next go to the line of Jeremy Kliewer with Deutsche Bank.

I know you've mentioned that Q2 call, you'll kind of give some more guidelines on Hawesville restart. But I was just wondering, what has been freed up? Or how have you found, I guess, the ability to expedite it by, whatever, 3 to 6 months? [At that time], it was a big issue.

It was just Gantt charting out literally. I wish you could see, I was in the plant for a couple of days last week, and they've got literally every cell in the plant Gantt charted out by the 9 steps starting with digging SPL out of a cell that's been curtailed and ending with putting a pot on [vat] and power. And so it was simply, we never like to, either internally or certainly to our investors, get ahead of ourselves. And so what we told you guys last time was what we briefed our board on -- the last time we met with our board about 1.5 months ago.

And it was simply rolling up our sleeves. And with a view towards, as I said, that we believe, in this environment, every incremental tonne has value to our shareowners. And so with a view towards getting those tonnes on as fast as possible, it was simply doing the work to see how quickly we could get those cells back online. There was no sort of single bottleneck that went away or project that we deferred or anything like that. There was no big bang there.

Okay. And then Shelly gave some great color on EBITDA kind of expectations in the upcoming quarter. I was wondering if you could kind of give the same kind of bridge or outlook for working cap for the rest of the year. If there's going to be a big drag on cash flow, or is it going to be a source of cash towards the end of the year? [If you can give] any kind of color there, that would be great.

Let me give you a little color. It's tough to really do any sort of forecasting, given that you've got to make an assumption about prices going into your inventory. But a couple of things to note. In the first quarter, we had that big build in inventory related to receivables that was due to selling more to direct customers with longer terms. That should -- we should be through that. That shouldn't recur. That was just building into the new contracts for 2018. That said, in the second half of the year, we are going to have to build some working capital for Line 5 and the other lines restart. So there will be some additional inventory, we'll have to build up. So you've got some offsetting factors then. And then it's really just a matter of what your raw material pricing does.

Let me just, if I may, also make a comment. It should be obvious, and I believe we've talked about it before on the direct sales. And so, as you know, our strategy has been, over the last couple of years, to drive more value-added sales, and we've succeeded in doing that. When we -- our strategy, from a distribution standpoint, is for commodity products to go through third-party intermediaries, traders, marketers. That's -- it's a standard product, and there's nothing custom that needs to be done in terms of touching the customer. It's the cheapest way to go to market. And we believe, and we believe it because it's what our customers tell us that on the value-added products, they want to buy from us directly. It's more of a bespoke sale, they're custom alloys. Different billets and alloys perform in our customers' plants in different ways. And so we want to go direct because our customer wants us to go direct.

The payment terms that we get from our trade -- main trader, purchaser of our standard products are very, very short. Very, very short. They pay us very quickly, in a matter of a couple of days. Our direct customers pay on industry standard terms, 30 to 40 kind of days. And so that's the difference there. As you would expect, the IRR of that investment, i.e., the incremental margin that we get by going direct and not through a trader on the value-added products divided by that investment in working capital, is a very attractive IRR or else we wouldn't be doing it. But just wanted to belabor the point that we're not simply investing in working capital to invest in working capital. There's a high unlevered IRR financial return by doing that, plus, of course, as I said, we're servicing the customer how they want to be serviced.

We'll next go to the line of David Gagliano with BMO Capital Markets.

First one, just on the timing on the Hawesville stage restarts here. Obviously, it's been pulled forward a bit. But just wondering, you also mentioned it could be stopped on a dime. What's the next day or week or whatever that we should be thinking about in terms of a go or no-go decision? And what alumina price do you need to see or expect to continue to push forward with the stage restart here?

Yes, that's a good question, David. in terms of -- I'll answer the last question first, and I might ask you for some clarification on the first part of the question. So as we told you last time, let me just isolate some math for you, we told you Line 5, the first line to be restarted, has an incremental EBITDA -- incremental investment [of system]. That hasn't changed. As I said, we're at least on budget or better than budget on the entire program, not just Line 5. And at the time that we told you, basis commodity prices at -- the spot prices at the time meant incremental EBITDA of $25 million. Now, at spot prices today, that number is lower, of course, because they -- increase in alumina has overshadowed the increase in the Midwest transaction price. But it's still nicely positive.

Just to isolate your question. At the current metal price, alumina would have to go into the mid-$700s for there to be no incremental EBITDA on bringing up Line 5. And so, right now, a 6-month return -- pardon me, 6-month simple payback that we told you about last time looks probably more like an 11- or 12-month simple payback. Even at spot prices today with alumina at -- priced at $640, again, that $640 would have to go to mid-$700s for the -- for it to go to a 0 incremental EBITDA. So I think that hopefully answers your second question. We've got some room.

We think there's -- I think those numbers show the attractive nature of the project because once alumina goes back into the below $500, below $400 current metal prices, you start talking about a simple payback in a matter of just a couple of months. And the same math applies for the full 3-line restart. Can you ask your first -- David, I apologize, your first question again -- the first part of your question again? Because I didn't quite follow it.

Sure. I was trying to figure out if there's any sort of day that you need to make a decision, not on Line 5, but the other -- I guess, the other 2 lines and...

I'm sorry, David.

No, that's okay. And also -- but related to that, you just mentioned one other thing. And I think I heard you correctly. Are the economics the same for the other 2 lines that you just mentioned for Line 5? Is that correct?

Yes, they're pro rata to what we gave you last time as to what was -- I just gave you now for Line 5 -- for the first line, Line 5, absolutely. On the first part, I now understand. I apologize. There's really no day. There's no like Big Bang day where we have to make large commitments. But the biggest commitment forward we make here is to order some of the long lead time materials, the cathode bricks and collector bars. But you're not talking about significant forward commitments there. Otherwise, literally, I wish you could -- you're more than welcome to come to the plant and see it. It's just -- it's labor, both internal labor, digging cells, et cetera, et cetera; and then rebuilding cells internally; and then external contractors, things like yanking superstructures and getting them straight; and all the process in that 9-step process about which I -- which I summarized in the Gantt Chart.

And so there's really not a -- there's truly not a big, Big Bang. We could -- it would be a terrible thing to have to do if the market really, really got sideways. Obviously, we're watching it closely. But it's not an exaggeration. We really could stop the project at any given time. That having been said, our intention is and sort of the default answer is to proceed apace. We think we're going to -- our investors are going to get paid if we do that.

Okay, that's helpful. Then just switching gears for a second. Earlier, I thought I heard Shelly mention $20 million of cash spending that will be capitalized related to Hawesville restart in the second quarter that will be capitalized, I think, you said. Are you still planning to account for about $95 million of the startup costs in operating expenses? Or have you decided to capitalize these?

Yes, that's a good -- that's a great question, David. And the answer is, with very near certainty, we're near the end of our analysis that we're going to be capitalizing as is the industry convention. As we've discussed before, through our research, we've determined we're the last -- with -- "primary aluminum company" to still expense pot rebuild. And in fact, under international standards, as we've done our research, it's not even permitted. You must capitalize and depreciate over the economic life of the cell.

So at least for the restart at Hawesville, and I underscore that, for -- you hit it perfectly. For that $95 million, our intent -- again, we're not 100% of the way there yet, but we're very close to 100% of the way there, and that's why Shelly talked about it as she did is to capitalize. We've not yet even considered, David, at all, changing our treatment for, I'll call it, the normal reline activities. This is -- we're basically rebuild -- not basically, we're rebuilding the entire reduction department at Hawesville, all 5 lines, all 560 cells.

And that's why we think it makes sense, and the economy experts think it makes sense to capitalize [at that rate]. It just wouldn't be right to expense it in a number of ways. Going forward, we'll have to study the situation in terms of normal pot reline expensing, but for now, no change there. But you -- one more time, you put it perfectly. That $95 million, of which $20 million, as Shelly said, will be spent in the second quarter, For all intents and purposes, it will be capitalized.

Okay. Then one last quick one. Shelly, thanks for the bridge to the second quarter, obviously, clearing it up and spelling it out. Just to round it out. Any other movers in the second quarter we need to be thinking about, for example, incremental carbon costs, power prices declining, that kind of thing?

No. The only thing that I would mention is, we've talked about that cold snap that we had in January that cost us about $6 million in the first quarter. We wouldn't anticipate something like that recurring in the second quarter. Other than that, no major movers.

Carbon looks pretty flat, David. That's been, as you well know, been the other one the last couple of quarters. So Shelly gave you the big movers. And as she said, we did eat a couple of million bucks, more than a couple of million bucks in Q1 due to that cold. And that seems to happen every 4 years or so. And thus far, that hasn't repeated, unless we have some strange weather or some strange transmission problems in the Midwest grid, that shouldn't repeat in Q2.

(Operator Instructions) We'll now go to the line of David Lipschitz with Macquarie.

A quick question with the whole Rusal situation. Are you impacted in any way from that in terms of alumina in any fashion, if there were to be sanctions come back in October?

No, I mean, other than the index price being where it is, obviously, we're very impacted by that. But otherwise, no. We did, from time to time, like a lot of people in the industry, just given its size, take alumina from the Aughinish refinery in Ireland. That's a Rusal refinery. It's an excellent plant. As you'd expect, it -- when we took cargoes from Aughinish, it would go to Grundartangi, our plant in Iceland. But other than that, from a physical supply standpoint, no. And I'm trying to think of any sort of downline. I guess, the answer is really no, again, other than the obvious impact on the price.

Okay. And just a quick follow-up. I just want to make sure I got it straight. When you said that the second quarter EBITDA was -- did you say it was going to be around $45 million or that's excluding like the addback or the power that you got hit with in the first quarter, that you have to add to that? Or I just want to make sure that's -- that what you're giving is guidance for the second quarter.

Yes, let's start from first...

Yes, so walking you from Q1, $22 million of EBITDA in the first quarter. We've got $15 million benefit from the lagged alumina price and then another roughly $25 million from prices, meaning LME and regional premiums. That $15 million and the $25 million gets you $40 million on top of the $22 million for this quarter. So you're in the low to [mid-6s -- mid-$60 million].

So she's just given you a bridge. And remember that in all these -- she's given you price changes. I should say, realized price changes. Important to understand, realized quarter-to-quarter.

And we're going to return to the line of Novid Rassouli with Cowen and Company.

I just wanted to touch on Slide 5. It looks like based on the figures, you guys are expecting a deficit of about 1.2 million tonnes for aluminum in fiscal '18. Is that correct?

Yes, according to the market experts, that is correct. This is -- this considers the restarts such as ourselves that we've announced and any supply changes in China or the Western world.

That looks very thin right now.

Yes.

Yes. So that's the one thing I just wanted to ask. I think there's kind of an expected 4 million tonnes of new capacity expected to come online in China this year. I just wanted to see if that number -- what kind of level of new capacity for China that number includes, if you guys have that off the top of your heads?

It's got to -- I can't verify the 4 million. It's a couple of million, definitely. Let's see. We're working off a base of [33 7, 8]. It's at least 3 -- it's got to be between 3 million and 4 million in there, just given the growth rates in production off the base -- off the '17 base. I can't precise that number, but you're in the right ZIP code.

That does exhaust all questions in queue at this time. Please continue.

Well, we thank you, as always, for joining us and look forward to talking with you when we report results for the second quarter. Good evening.

Ladies and gentlemen, that does conclude your conference for today. Thank you for your participation and for using AT&T's Executive TeleConference service. You may now disconnect.