Century Aluminum Co (CENX) 2017 Q4 法說會逐字稿

Ladies and gentlemen, thank you for standing by, and welcome to the fourth quarter 2017 earnings call. (Operator Instructions) And as a reminder, today's conference is being recorded.

I'd now like to turn the conference over to your first speaker, Peter -- excuse me, Peter Trpkovski. Please go ahead.

Thank you, Ryan. Good afternoon, everyone, and welcome to the conference call. I'm joined today by Mike Bless, Century's President and Chief Executive Officer; and Shelly Harrison, our Senior Vice President of Finance and 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.

Turning to Slide 2 of today's presentation. 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, Pete, and thanks, everybody, for joining us this afternoon. We know it's a busy week, so let's get right to it.

If we could turn to Page 4, please. I'll take you through a high-level overview of the last couple of months. We're pleased with the company's performance in the fourth quarter and also into the early part of 2018. Safety performance was good across the board with most plans improving quarter to quarter. Our focus on the identification of hazards and the prevention of life-altering events and significant injuries continues to pay dividends for us. At Grundartangi, we've seen a reinvigorated focus on the entire safety culture in their systems and processes that back it up, and this from a base of an already strong safety environment at that plant.

Plant operating metrics reflected the stability of the operations throughout the quarter and again, into 2018. And this, coupled with very good management of controllable costs, produced strong financial performance.

If you've had a chance to look, you've seen adjusted EBITDA was $60 million for the quarter. And this includes some impact of raw material price increases as well as increased logistics cost due to the continuing problems on the Ohio River. If you remember, we talked about situation last quarter.

Cash flow was strong other than the impact of the purchase of raw materials at much higher prices during the fourth quarter. This will go the other way beginning in the first quarter of this year, and Shelly will explain that.

As we predicted in October, we've seen a meaningful fall in raw material prices over the last few months, though we're still not quite the back to what we would consider to be a normal environment. In Q1, we'll see the worst impacts of these raw material prices from an income statement perspective given the lag in realized pricing, as we've discussed with you many times.

Shelly will give you some more detail on the commodities, but let me just give you a quick summary now. As you'll recall, alumina had been trading in the low $300 per tonne for most of 2017. And then in the early fall, we saw a really rapid increase to just shy of $500. In fact, it was just shy of $500 when we released our results in October, as you'll remember. At that time, we said we believe the run-up was due to some nonfundamental short-term imbalances. And we've now seen the price return most of the way, now trading in the mid-$350s.

Same story with calcined petroleum coke. The delivered price had rocketed up from the mid-$300s to around $500 in the fall, and we now see the market price around $400. Like alumina, we believe coke still has some way to go to return to a rational price on a fundamental basis.

Pete will give you some more detail on the industry environment in just a few moments, but let me just give you a couple of quick high-level comments. The Western world outlook continues to be favorable in our view. Demand growth remain strong. And barring some kind of global macroeconomic shock, we'd expect this environment to continue.

The picture in China is not nearly as clear nor as favorable. The consensus is clear that the capacity cuts have not curtailed as much production as expected. Again, Pete will give you some more detail on that. Further on the supply side, it's now clear that China is looking hard at building capacity in other parts of the world, principally in Asia.

On the demand side, the picture is also somewhat troubles. We see China domestic demand growth coming down from the double digits to the mid-single-digit range in 2018. The impact of the situation isn't hard to predict. We believe exports from China will set another record in 2018. This is why we believe it's absolutely critical that the final order under the Section 232 investigation be made immediately.

Flows into the U.S. have increased markedly over the last year, and we believe it's clear that this is due to attempts to get metal into the U.S. before any market adjustment. In addition, we believe there are now well over 0.5 million tonnes at foreign ports being loaded on the ships, and this material will be in the U.S. in a matter of weeks.

We're encouraged by the Commerce Secretary's report recently sent to the President. The report, if you've had a chance to read it, send clear harm for the U.S. national interest -- national security caused by state-subsidized aluminum flowing into the U.S., which is absolutely consistent with our fundamental analysis, as we've described to you many times. The objective of the proposed remedies in the report is also consistent with our thinking. This objective is to adjust the market so that the available U.S. capacity can restart.

We're still studying the alternative remedies proposed in the report, and we look forward to working with the administration on finalizing this process. Again, I need to reiterate, speed is absolutely critical in our strong view. Any final remedies are going to be starting with a mountain of recently imported material, putting a significant drag on market conditions.

Assuming these market abuses and the resulting imbalances are corrected, we increasingly believe the U.S. will be an attractive market for primary aluminum production. Obviously, the market is hugely undersupplied. We have growing vibrant downstream industries and attractive cost position in terms of wholesale electric power.

In this environment, we believe the company is really well positioned. Our plants are in the right places. We're making the right products today, and we continue to make further investments in our value-added products capabilities. We're absolutely convinced we have the best technical and operations talent in the industry, and we're comfortable that the U.S. will maintain its advantage in wholesale electric power prices. So we're excited about the opportunities we've got.

Given this, we have the confidence to begin the first step of setting up our partially curtailed U.S. plans for what we believe should be long lives. Our focus, of course, is on Hawesville. As a reminder, the restart of the second potline at Mt. Holly remains dependent on solving the power issue there, and I'll make some comments about the status of that in just a moment.

So now back to the Hawesville. As you know, we curtailed 3 of those potlines, 3 of the 5 potlines at Hawesville, in late 2015 as the price came down precipitously. Since then, we've been running the plant to minimize any discretionary spending, obviously, maximize cash flow.

In times of normal operations, as those of you have been around smelters now, you're regularly rebuilding cells when they fail after their normal period. But the Hawesville cell usually last about 5 years. So in a normal year, you'd be rebuilding about 1/5 of the plant cells every year, and that costs us about $20 million a year. We've deferred this activity since we closed the 3 potlines. We've been cannibalizing the available cells in the 3 closed lines, and we've best been avoiding that annual investment.

As we told you before, we're now near the end of our ability to do this. We think we could probably limp through 2018. However, both currently producing lines will need to be rebuilt by the end of 2019. Again, that's just to maintain the current production of about 100,000 tonnes a year. We're seriously considering rebuilding one of those lines this year. That's to spread the investment over the 2 years, i.e., do 2 lines over 2 years. That's to more evenly balance the work required in the plant given just the physical constraints in the plant. Again, one more time, this is simply to maintain the current production level.

At the same time, we've been assessing the installation of a new pot technology. We've got several cells with the new technology now installed in the line -- the 2 lines that are operating, and the initial results look very promising. The technology results in more metal and a lower power usage in each cell. And we're likely to make the decision in the coming months whether to go with the old or the new pot technology when we start rebuilding cells.

We're also preparing for the potential decision to restart the 3 curtailed potlines at Hawesville, and that decision will be based on our level of confidence that the 232 action will produce a rational market environment. We're putting ourselves in a position to make and implement that decision very quickly. Assuming our confidence in the 232 remedy, it's going to be a reasonably easy financial decision. The marginal EBITDA in the restarted potline quickly repays the investment and the restart costs, and I'll give you some detail on those -- on both the costs and incremental EBITDA in just a couple of minutes.

Just quickly, as promised, an update on the situation in South Carolina. Again, the decision to restart the one curtailed potline there, its 1 of 2 curtailed potlines there, is based on getting access -- full access to market power, not based on the metal price.

As a reminder, for the last several years, we've been buying about 3/4 of the power requirement from the market. And we've also been forced to buy the remaining 25% from the local monopoly power supplier. The price of the market power itself, that 75%, is consistent in what we pay in Kentucky. The only difference is in South Carolina, we're forced to pay 2 transmission charges, one to the market power supplier to transmit the power to the South Carolina system border and another to the local utility to transmit the power to the plant.

If this were the only issue, we could likely deal with it. The delivered price would only be about 10% higher than in Kentucky due to that double transmission payment. However, as we've discussed many, many times, the price of the power from the local monopoly supplier is in excess of 2x the price of the market power.

On 100% market power, Mt. Holly would be solidly in the second quartile in the global cost curve, and that's due to its excellent work for us in modern technology. Instead, the current weighted average power price, 75% market power, 25% local power, that weighted average yields are worse than median cost position.

We're again hard at work trying to engage the decision-makers in South Carolina. We've made a formal written offer that would enable us to restart that idle potline. In our offer, we'd buy 100% of our power from the market sources. Of course, that's power sufficient to run the entire plant, both potlines. We'd continue to pay the local power company at standard transmission rate per megawatt hour as we're paying today. We'd also pay an annual access fee to the local power company to ensure that the contributions they receive from us in the future toward their fixed costs would be equal to what we're paying them today.

We firmly believe the facts support the logic of this proposal. The power companies made whole. The second potline at Mt. Holly is restarted, and several hundred jobs are restored. And the state benefits from an incremental $0.5 billion in annual economic activity. And we're presently doing everything we're able to move this forward expeditiously.

And with that, I will turn it over to Pete to talk about the industry environment.

Thanks, Mike. If we can move on to Slide 5, please. I'll take you through the current state of the global aluminum market. The cash aluminum price averaged approximately $2,100 per tonne in Q4, which reflects a 4% increase over Q3. The aluminum price on a 2-month lag basis was up quarter-over-quarter almost 10% and averaged $2,087 per tonne. Since the beginning of 2018, aluminum prices have averaged approximately $2,200 per tonne and are currently sitting right around that level.

In the fourth quarter, regional premiums averaged approximately $0.095 per pound in the U.S. and $160 per tonne in Europe. However, spot premiums are significantly up and are currently approximately $0.145 per pound in the U.S. and $170 per tonne in Europe.

In the fourth quarter of 2017, global aluminum demand grew at a rate of almost 6% as compared to the year-ago quarter. We saw 7% year-over-year demand growth in China, 4% growth in Europe and 2% growth in North America. Global production growth was up 3% in Q4 versus the same period last year, driven almost entirely by increases in production in China, which increased 5% year-over-year despite the announced capacity curtailment program. As a result, for the full year 2017, the global aluminum market recorded a modest deficit of approximately 90,000 tonnes.

Since the U.S. has taken specific action of filing the WTO trade case and launching the Section 232 investigation last year, we have begun to see some initial positive supply side responses from China. The first response, announced shortly after the WTO case was filed in January, is the winter heating season curtailment program. Under this program, aluminum producers as well as alumina and petroleum coke producers in certain provinces in China were required to curtail 30% of their production during the winter heating season.

As a result of this action, we estimate that approximately 1 million tonnes have been cut during the winter heating season. This was less than originally expected as enforcement of these cuts was lax. There's yet to be any confirmation whether these cuts will be implemented again in the 2018 and 2019 winter season.

The second response, announced shortly following the initiation of the Section 232 investigation, required the curtailment of illegal, unpermitted production in several provinces. In order to restart this curtailed capacity, producers will need to obtain new licenses or purchase licenses for curtailed legal capacity. As a result of this action, approximately 4 million tonnes have been cut.

While these cuts were incrementally helpful, China still expanded in 2017 and will expand again in 2018. With a China's expansion in 2018, they will be in a surplus of 1.5 million tonnes, while the rest of the world will be in deficit of 2.3 million tonnes. As a result, we expect a global deficit of nearly 800,000 tonnes in 2018.

On January 23 of this year, the U.S. Commerce Department submitted their 232 recommendation to the President. The President then has 90 days to decide on what course of action to take. Just last week, Secretary Ross outlined the proposed recommendation. As Mike said, we must take swift action before that metal finds its way to U.S. shores before implementation of any remedy.

State-owned enterprises throughout the world including, but not limited to, China continued to illegally subsidize aluminum production in their home countries and export the problem to the U.S. The U.S. industry needs broad and comprehensive relief from Section 232 to address this issue and to allow the industry to get back on its feet.

With that, I'll turn the call back over to Mike.

Thanks, Pete. If we can turn to Slide 6, please. I'll just make a couple of quick comments about the operations and then let Shelly take you through the quarter and the year.

Starting with safety. Obviously, as always, as I said, we're satisfied with the company's safety performance this quarter. Mt. Holly and Grundartangi had terrific quarters and into 2018, no recordable safety incidents between those 2 plants. We saw good quarter-to-quarter improvement at Hawesville. As you see, a slight downturn at Sebree, but I'd note that plant is still at very good levels and had just an outstanding full year 2017.

Turning to operating performance. Hawesville has had a strong last couple of quarters. And you saw last quarter, quarter-to-quarter, nice production increase. That was on improving operating metrics. This -- the way the plant's been performing so well recently is an important underpinning of any decision we make to begin restarting capacity there. So we're really pleased to see that. It gives us a really good base to go forward.

Sebree continues to operate in a consistent and stable manner. You see incremental growth in tonnage there on stable operating metrics. And similarly, Mt. Holly and Grundartangi both consistent operations during the quarter and into 2018.

A couple of comments on conversion cost. Generally favorable across the plants, as you can see. What we're looking at here is really good management of controllable costs offsetting some very significant increases, largely in carbon costs. I'll take you through that in a moment. Remember, these are conversion costs, of course, so they exclude alumina. And Shelly, just in a couple of moments, will comment on the impact of the alumina costs in both Q4 and our forecast for the first quarter of this year.

A couple of comments just to give you a sense of the extent of the increases. As you can see, Hawesville's overall conversion cost improved a couple of points. That was in the face of a 44% increase in carbon costs. Same story at Sebree, flat in all, offsetting a 30% -- or against, I should say, a 30% increase in carbon costs. Mt. Holly, same thing, cost down a little bit in the face of 21% increase in carbon costs.

And lastly, at Grundartangi, you can see cost at 4%. If you take out the impact of the power price increase, of course, as you know, our power price there is 100% linked to the LME price. So it's really just an LME price impact. If you took that out, it would be a 2.5% increase. And all of that 2.5% increase, 2.5%, pardon me, increase was due to carbon on the one hand and increased potline expense on the other hand.

And with that, I'll give you over to Shelly, who will take you through the quarter and the year. Shelly?

Thanks, Mike. Let's turn to Slide 7, and I'll take you through the high-level results for fourth quarter and the full year. On a consolidated basis, global shipments were up 2% quarter-over-quarter, reflecting a 2% increase in production at Hawesville as well as some impact from timing of shipments at our other facility.

Looking at operating results. Adjusted EBITDA was $50 million this quarter, and we had adjusted EPS of $0.26 a share. Adjusting items for Q4 included a $7 million noncash gain related to the termination of legacy contractual obligations. We also had a $3 million noncash charge for lower cost to market inventory adjustment and a $7 million adjustment related to the final settlement of our 2017 LME hedges.

Turning to liquidity. Our cash balance remains relatively flat, and higher EBITDA was offset by significant investment in working capital. The working capital increase was driven by extremely high raw material prices at the end of the year, and I'll talk about that more in a couple of slides. Availability under our revolving credit facility is increased by $23 million on the back of the higher working capital balances that I just mentioned.

Okay. Let's go to Slide 8, and I can walk you through our Q-to-Q bridge of adjusted EBITDA. During Q4, we produced $60 million of EBITDA as compared to $48 million in Q3. The $12 million increase was driven by a $30 million improvement from LME and regional premiums, partially offset by the $21 million in raw material price increases that we forecast on our last call.

Alumina cost for Q4 were based on a realized undelivered price of $338 a tonne, which is in line with the 3-month lag index price. This is up significantly from the Q3 realized price of $296 a tonne. However, alumina prices continue to increase dramatically into Q4, and we expect the realized alumina price for Q1 to be around $445 a tonne. Alumina prices have come off their highs and are now sitting just below $360 a tonne. But we won't see the P&L benefits of these prices until Q2.

On the carbon side, you'll see a similar story with price increases impacting our Q4 results but even more so in Q1. It's important to note that from a cash flow standpoint, the cash lag is much shorter than the accounting lag. It's why you won't see the P&L impact at Q4's higher raw material prices until Q1. The cash outflow has already occurred and you'll see that on the next slide.

Okay. Just to lay this out for you like we did last quarter. We expect the impact from higher realized alumina prices in Q1 to be around $40 million. And for carbon, we expect this to be about $10 million. We also had a couple of cold snaps early in the year that caused our Kentucky power prices to spike for a brief period. We expect these higher power prices to impact Q1 EBITDA by about $4 million. Temperatures have normalized now, and we're seeing Indiana Hub prices nicely back in the 20s on a per-megawatt-hour basis. From a top line perspective, LME and regional premiums are expected to improve Q1 EBITDA by about $10 million.

Okay. Let's turn to Slide 9, and we'll take a quick look at cash flow. We started the quarter at $174 million in cash and ended the year at $167 million. Capital expenditures during the quarter were $8 million, and we paid about $7 million for LME hedge settlements. In Q1, you'll see there are $2 million cash impacts for the final payout on our 2017 LME hedges, but there will be no P&L impact associated with these hedges in Q1.

At this point, only some very modest hedge volumes are outstanding related to years 2019 and 2020. In December, we also made a semi-annual interest payment of $9 million. But by far, the biggest cash outflow during the quarter was driven by the investment in working capital that I mentioned earlier.

During the fourth quarter, we invested roughly $40 million in working capital, primarily in the form of inventory related to increased raw material prices. We're seeing prices return to more normal levels now, and we expect to see some of this investment come back to us in Q1.

Okay. Let's turn to Slide 10, and I'll hit some highlights for the full year. Year-over-year revenues were up $270 million, an increase of 20% on the back of higher LME prices and regional premium. Shipments were relatively flat with Mt. Holly and Hawesville partially curtailed for the past 2 years. Despite significant raw material price increases, we were able to turn 50% of our sales increase into EBITDA. Year-over-year adjusted EBITDA is up $135 million from $29 million in 2016 to $164 million for 2017. For the bottom line, we saw adjusted net income increase by roughly $100 million, which translates to more than $1 per share improvement in EPS.

And with that, I'll hand it back to Pete to talk about 2018.

Thanks, Shelly. If we can turn to Slide 10 -- excuse me, 11. I'll take you through the company's expectations for financial measures in 2018. Sebree and Grundartangi continue to run at full capacity, while Hawesville and Mt. Holly run at 40% and 50%, respectively. As Mike said, we are getting closer to a decision on rebuilding pots of our existing production at Hawesville, where we have been cannibalizing pots and deferring pot realigning spend. In addition, a decision on a potential restart of Hawesville's curtailed production could be coming soon. Until either decision can be made, we have not yet included any deferred cell realigning costs or restart costs in these 2018 items.

As many of you know, our selling price is comprised of LME price, regional premiums and value-added product premiums. We give you the tools to sensitize for your own LME and regional premium in the appendix of our presentation.

As in prior years, we give you our expectation for the premium we receive on value-added products over a standard grade aluminum. We estimate approximately $190 per tonne over the LME and regional premium on average over just our value-added tonnes, not weighted average overall tonnes.

Now moving on to our key cost components and cash costs. We've broken out our cost between Q1 and Q2 to Q4 so you can see the impact of the lag accounting in Q1 versus our expected performance for the rest of the year. As Shelly discussed, Our Q1 cost will reflect the extremely higher raw material cost we saw towards the end of last year and higher power prices in the U.S. so far this year based on a couple of cold weeks in January. Our Q2 to Q4 cost reflects more current levels.

You will notice our gross plant cash cost from Q2 to Q4 are still up in the U.S. and Iceland from our prior year items. This increase is more than 100% related to alumina and carbon price assumption in the U.S. and about 70% in Iceland. The remaining increase in Iceland is related to power as it is linked to higher LME prices.

We also present our cash cost net of all premiums received above the LME, but this metric is directly comparable to the LME price. If you take the LME and deduct our net cash cost, the result is our expected cash margin per tonne with no further adjustment needed. You can find our underlying assumption and reconciliation of our net cash cost in the appendix of today's presentation.

Okay. If you can turn to Slide 12. I'll give a couple more comments before we turn it back over to Mike. Our SG&A expense is expected to be $43 million, $8 million of which is non-cash. Moving down the CapEx, similar to last year, our expected spend is between $25 million and $30 million, of which $15 million to $20 million is related to maintenance.

On taxes, we continue to expect our U.S. NOLs to shelter essentially all of our U.S. taxable income other than some modest state taxes. In Iceland, we will continue to accrue a rate of 20%. As you would expect, we did a full analysis of tax reform at year-end and ultimately determined that the impact on Q4 and going forward is expected to be minimal due to our large U.S. NOLs.

Lastly, our consolidated cash flow breakeven using all of the items just discussed is $1,875 per tonne. As a reminder, this was an LME direct equivalent number and represents our cash flow after maintenance CapEx, SG&A, cash interests, cash taxes and any other corporate cash outflows but excluding any discretionary CapEx. Just like our plant cash costs, the increase from our 2017 expectation on items can be more than explained by higher alumina and carbon prices.

With that, we can now turn the call over to Mike for his closing comments.

Thanks, Pete. We want to get right to your questions. Just a couple of last thoughts. As I said, we're really constructive about the future of the company at this point, including especially the U.S. operations. As Pete and Shelly both summarized, Q1 will be an anomaly as we've got the higher vital costs running through the income statement. To reiterate again what Shelly said, the cash for this has already been spent in the fourth quarter.

Pete will give you the cost structure after the first quarter. As he said, those estimates still include what we believe to be abnormally higher raw material pricing. We wanted to come in on the conservative side there. In addition, we have some investments to address, as I said, 2 years of deferred maintenance and other spending in the U.S. plants.

If you take a step back, you probably haven't had a chance to work with these estimates yet. But if you take these, the cost structure that Pete took you through and the other estimates, and you were to use current spot prices both for costs, commodities, of course, and LME and premiums, you'd get an annualized EBITDA, just to give you a sense, of around $300 million. And you get the same answer if you took Q4, adjusted it for spot prices versus the realized prices that we had in Q4 and put in the increased investments for the catch-up deferred spending at the plants. As Pete said, this is before any cell rebuild activity at Hawesville.

In that respect, let me just walk you through quickly what the economics of a potline restart at Hawesville would look like. Now I'm focusing on these 3 curtailed potlines. So the first line, if we made the decision within the next couple of weeks, we could have the first line producing the first pots on power no later than the end of the first quarter. And then by the end of the second quarter, maybe even a month before that, but let's call it by September -- pardon me, sorry, end of the second quarter, to have the first pot on power, pardon me, and end of the third quarter to have the potline in full productive capacity, i.e., the incremental 50,000 tonnes.

So that's the first potline that we would do. The cost of bringing that back on is about $15 million. That's primarily operating expense, cell rebuild. As you know, we expense that immediately. We don't capitalize it. And there's a little CapEx in that as well. So that's $15 million, 1-5 million, is a combination of OpEx and CapEx. We'd rehire 90 folks. There would be an incremental 50,000 tonnes, as I just said, of incremental production. And the incremental EBITDA, once -- on an annualized basis, once the plant was at -- the restarted potline was at full capacity again, around the end of the third quarter, at current spot prices, would be in the range of $25 million to $30 million.

So a $15 million investment, $25 million to $30 million of incremental EBITDA. So you can see, you're looking at about a 6-month payback. And now you can hopefully understand why at the beginning of my comments I said that assuming the 232 order gives us confidence that the market will be rational, these decisions to restart the potlines at Hawesville are reasonably easy ones from an unlevered IRR standpoint. So that's the first potline.

And then the further 2 potlines, we're just completing the analysis on those. The restart costs on the fourth and fifth potlines, or I should say the second and third currently curtailed potlines, is a little bit more than the restart cost in the first potline. The way the plant works is that if you're running in excess of 3 of the 5 potlines, you need to make certain investments, mostly capital investments in the plant's infrastructure to bring some of the support department back up.

So the investments there will be slightly larger, but the incremental EBITDA on each of those lines will be slightly larger as well because you're further leveraging the fixed cost of the plant. And so the paybacks there are going to be perhaps a little bit longer than 6 months but a simple payback, pardon me, well under a year. And again, from an unlevered IRR standpoint, you can do the math in your head, would still make quite a bit of sense.

So we're excited about all of that, and we hope to be -- as soon as we have a look at the final order that the President comes out with over the next 1.5 months, we hope -- as we've said, I hope we've been clear, sooner rather than later, we were looking forward to talking with you again about our decisions to move forward with all that.

And with that, we'd like to turn it back to the operator to take your questions. Just a request to bear with us a little bit. I'm in a different location than Pete and Shelly. So please bear with us if we fumble a bit as we decide who's going to answer your questions.

With that, Ryan, we can now -- we can get going.

(Operator Instructions) Our first question will come from the line of Novid Rassouli with Cowen and Company.

So Mike, on the restart of Hawesville, can you just walk us through as far as maybe incremental demand for high purity relative to maybe non-high purity aluminum? And how much do you think the market could absorb of that if we do get something positive on the Section 232? And perhaps maybe -- I don't know if that had to do with the fact that potline 4 and 5 have incrementally more EBITDA or not, but if you could maybe help us frame that as that's definitely been, I know, a stress of you guys in the past several months as far as Section 232 and I think Wilbur Ross actually stressed it as well in his recent conference call?

Yes, that's a great question. Thank you. And I probably neglected to point out something Secretary Ross has discussed many times, including in his press conference on Friday. So the first is a factual point. In that incremental EBITDA that we gave you in that calculation, there's no incremental purity assumed. We want to be conservative as to what the incremental product would be. And so we think there may be some purity demand incremental that we could get later this year, and we might eventually put an element of that in it. But generally, there's very little, if any, purity assumed in those numbers. Number two is that on a broader scale, in the real world, we do think, obviously, Hawesville can ramp back up to up to 100,000 tonnes of annual purity production. This is 0404 and better. A large portion of it is 0202 and 0303. And again, you cited it correctly. A significant component of the Commerce Department's spot set in the recommendations to the President, as you read in that report, have to do with preserving the high purity capacity at Hawesville, which, of course, is the only purity producer in volume in the U.S. So we do believe that going forward, assuming that the market is adjusted appropriately, that we will have opportunity to re-enter the purity market. But we didn't want to make a lot of assumptions that, for example, a whole potline or a majority of a whole potline would be able to capture purity immediately. The other thing I would note is that from just a technical perspective, it'll take another couple further months, not many, but another month or 2 to make sure that the pots are in appropriate operating configuration to make the purity. You need to a really, really stable potline especially to make the 0202.

Very helpful. And then just sticking on that, would you be able to comment on the incremental EBITDA above and beyond just non-type purity aluminum relative to the high-purity lines? And then what percentage of the market currently is served by imports for high purity? I'm just trying to get a sense of what the opportunity is here for you guys in the future.

I'll answer your last question first because it's an easy one. 100%. We're not making any purity today. So after the -- and we haven't since -- let's see, we haven't for the last 2 years, and so after -- or 21 months, I would say. So after the market was saturated with product from 2 regions in particular, and these are called out, of course, in the Commerce Department's report, the Persian Gulf and Russia, at well, well, well below established market prices, we stopped purity production at Hawesville. It didn't make any sense to us. So the answer is 100%. I'm not sure, Novid -- I'll try to answer what I think you were asking in the first part of your question. But you redirect me please or come back if I'm not exactly on point. So that incremental EBITDA, again, assume very little incremental purity, just a smidge. As I said, Hawesville has proven that it can produce up to 100,000 tonnes. So if you assume that we did bring up the fourth and fifth potlines, which again is our strong intention, assuming the 232 order makes sense to us and can correct the market, we believe that there could be a good chunk of that second 50,000 tonnes and the third 50,000 tonnes that could be high purity. As I think you know, I mean, the typical market over time for 0202 has been well, well, well over $200 a tonne, approaching $300 a tonne and more. We make a little 0101 as well, which can be $500 a tonne to $800 a tonne. And even 03 and 04 have traded $0.04, $0.05, $0.06 -- I'm sorry, $100, $150 a tonne. So there's good incremental opportunity there for just pure incremental cash flow. As we told you in the past, it doesn't cost us significant additional operating expense to make the purity. You just need to man those potlines with experienced people who know how to tend those cells, and you are limited somewhat in your alumina choices. There's maybe 6 or 7 or 8 alumina to which you're limited. But that won't narrow our current supply base at all. So I'll stop talking now, and you tell me if I got it or didn't.

That's great, Mike. A very thorough answer. Appreciate it.

And our next question will come from the line of John Tumazos.

Has Ravenswood been bulldozed? Could it be brought back? Or with appropriate regulatory reform, could there be a possibility of a new smelter in the U.S. as opposed to Iceland?

That's a great question, John. So the sad answer is I don't know if Ravenswood has been bulldozed. Maybe my colleagues -- we sold it. Oh gosh, it closed a year ago or 2 years ago. I can't even recall now. And to my knowledge, although I haven't followed the situation closely, the buyer intended to use -- to, in fact, bulldoze, as you say, take the plant down and use the site for a different industrial purpose. Pete or Shelly, do you know? Because I don't.

No. Nothing specific. You were right. It was a year ago, it was, when we sold it.

Yes, okay, January of '17 then. John, in answer to your -- the second part of your question, we're hearing more and more talk about just that now. I think it's interesting to note that you hear a lot about power prices and a lot of people, especially going back to the 232, who say -- people obviously on the opposing side of this who say why support an industry that can't be competitive. John, as you know, you follow the industry closely, and you've seen all the data from the industry "experts" and consultants out there, the U.S. average wholesale electric power price now is about 10% below the world median. And so by no means is the U.S. disadvantaged in power prices. And so I guess a couple of years ago, we would deem each other crazy to even be having this discussion, I suppose. But now I guess you could envision with power prices where they are, if someone were willing to -- of course, as you know, John, it takes a couple of decades, 15 to 20 years, to earn back the investment on a brand-new greenfield -- that's redundant, on a greenfield smelter. So you'd have to have a power supplier who is willing to fix the price based on current prices. But as I think you know also, I'm going to stop my answer here in a moment, forward power prices to the extent the forward curves go, forward streams to go out, are flat or even in a slight backwardation. 5, 6 years from now, you can buy [Indy] Hub power for the same price where you can buy it tomorrow. So that's a long-winded answer of saying we don't know of any efforts that are actually an information, but we hear a lot of people talking about it.

I was thinking, Mike, $3, $3.50 long term for the gas and $0.04 or less for electricity, Mike.

Yes, John. I would say for an operating smelter on the one hand, clearly, or a partially curtailed smelter where you're going to bring back some capacity, something like a low-30s power price, like I said, where we need to get to at Mt. Holly if we're in -- at full access to market power payment, 2 transmission rates, that [dog hunts]. I would say, based on our calculations and expected returns and all that kind of good stuff, to build a greenfield plant, I think you need something closer to spot gas prices or even a bit below, kind of like mid-$2 gas and kind of high $20s power before you get to the kind of IRR that's going to get people interested on a couple of billion-dollar investment. That's what our math says.

(Operator Instructions) It looks like we have no further questions in queue.

Okay. Then thank you, Ryan. We very much appreciate everybody's interest and time this afternoon. And again, we look forward to speaking to you again...

Mike, sorry to interject, I just saw on the line here, operator, Ryan, I think we have another question just queue in for the last minute.

Someone beat the buzzer.

And from Macquarie, we have David Lipschitz.

Just a quick question with regards to the 232. How do you guys feel about investing if the government can peel it back at pretty much any time? Like how do you look at that from that perspective, that if a new President comes in or they decide to change things up in a year or 2? How do you work around that?

Great question. And to your point, they can change it anytime. I believe the Commerce Secretary was asked that question, if I recall, during the press conference. It might have been in a different venue. So they can change it anytime. We would feel confident if the initial remedy make sense to us, David, because to us, they clearly get it. If you read the report, as I said, it sounds like you have, their thesis is in line with ours. The objective of the remedy, whatever the remedy -- whatever structure they choose comes down is in line with ours. And what they've said, and we take them at their word, is that the only reason they would ever change it or withdraw it is if in their strong opinion the market have been adjusted successfully, i.e., the conditions had been created for the U.S. industry to be viable and competitive, not just viable over the long term. And so we believe they've got it right thus far, and we would, in essence, put ourselves in their hands that if they determine that things could change and they changed it or withdrew it or whatever and then the situation reversed again, they would take further action. They've taken strong action. It took longer to get to than we might have liked, but that is what it is. These things are complicated, and we get that. And so in terms of a new administration, that's not something that we can even think about. We're happy to be working with this administration, and we're going to be happy to bring this capacity back on as soon as we get that order that's in line with what was in this Commerce Secretary's report.

And then quick -- just quick last one, and maybe I missed it during the call. You gave a nice [look-through] with your cost for 2018. Is there anything you have for the cost for '17, what they were versus -- just a comparison of '17 versus '18?

Let's see. Pete, you want to take that one? I could comment, but you go ahead. And there's nothing in that format, I guess, Pete, but do you have for David off-the-cuff sort of quick guide as to how he might go about it?

Sure. And maybe, David, you can tell me specifically what you're looking for. LME, as I said, on a 2-month lag basis, was up 10% for the quarter. But for the year, the 2-month lag LME was about $1,909. For full year '17, premiums, 2-month lag -- I'll do everything on a 2-month lag basis because that's what's with the results, about $83.25 per pound in the U.S. The European Duty Paid Premium 2-month lag was $143 per tonne. That's on the revenue side. On the alumina side in 2017, you can go to the 3-month lag, but you're talking $320, $330 per tonne. Power prices, about -- just a hair under $30 for U.S. Midwest, hair under $3 per MMBtu for natural gas. And coke and pitch prices were significantly less in 2017 versus our 2018 guidance, as we've been saying.

So Pete took you through all the big assumptions, and you can take a look at what we have in the appendix to the presentation, and you can see how it compares to the 2018 assumption. And then we've also got the sensitivities, and that will give you a sense of how the 2017 numbers match up with what we have for '18.

Yes. I think, David, what you're going to find, what you're maybe after, back to the comment that Pete made in his remarks, is that the costs are up, no doubt about it. They're up because of the increase in commodities, alumina and coke and pitch, carbon, as we say, in jargon, and the incremental investments about which I talked in catching up on some of the deferred maintenance that we've been avoiding just to keep the plants going. But all of that, as Pete said, more than all of that increase is explained by the commodity cost increases. So we think we're doing a reasonable job of offsetting those, enabling us to invest in the plants and even catch up in some of the investments and still keep things going. I would say, again, we've reflected current commodity coke and pitch and alumina prices. I've just -- obviously, they could continue -- they could go back up, but our view is they're going to continue to fall. We've put in basically the spot prices just to err on the conservative side. We would hope to be able to take those down as the year progresses, but time will tell, of course.

And it looks like we have no further questions in queue.

Okay. Again, thanks, everybody. As I was saying, we look forward to talking with you when we report the first quarter and even more optimistically, hopefully, when we see a copy of the President's final order over the coming weeks. And we'll let you know if we have something to say at that time. Thanks again for your time.

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