Century Aluminum Co (CENX) 2017 Q2 法說會逐字稿

Ladies and gentlemen, thank you for standing by, and welcome to the Century Aluminum Company's Second Quarter 2017 Earnings Call. (Operator Instructions) And as a reminder, this conference is being recorded. I'll now turn the conference over to your host, Peter Trpkovski. Please go ahead, sir.

Thank you, Kathy. 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. 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.

Thank you, Pete, and thanks to all of you for joining us this afternoon. If we could just skip to Page 4 please. I'll give you a quick rundown of what we've been up to over the last couple months.

First, operating and financial results for the quarter if you've had a quick moment to take a look. They were in line with our expectations, and we believe showed very good progress quarter-over-quarter. Production metrics were stable and production volumes and conversion costs were favorable. And I'll provide some more comments on the operations -- pardon me -- in just a few moments. The operating performance led to strong financial results as you've had a chance to take a look, hopefully you've seen that EBITDA was up sequentially $12 million at Q1 to Q2 on a $23 million revenue increase. The results came in precisely as we expected. Let me just take you through the drivers at a high-level and then in a couple minutes Shelly will give you some detail.

There were 2 significant drivers this quarter that drove the results Q1 versus Q2. On the revenue side, obviously, you saw an improvement in the LME and in the regional delivery premiums. And on the cost side, we saw a meaningful increase as we expected in our realized alumina price. First on the metal pricing, obviously, the improvement #1 in the LME, #2 in the Midwest premium here in the U.S., and #3 in the duty-paid premium, all 3 of those together added $30 million of EBITDA Q2 over Q1, that's relatively straightforward.

On alumina, there seems to be some continued misunderstanding of how our reported costs work. So let me just take a few minutes here and go through it. We covered this last quarter, and I think it's worth just going through the assumptions and the math.

First as we described, our alumina contracts referenced the global index price on a 1-month lag basis, so we pay the price 1 month in arrears. Then, obviously, we have inventory on hand at all the sites and at any given time of course there's inventory on the water. We use FICO accounting as you know. And so when you mix all those together, that ages the inventory pricing by a further month and a half to 2 months. Again, we talked about this last quarter.

So the alumina price that rolls through our cost of sales each month and each quarter obviously reflects what we paid for that same alumina about 3 months ago.

Now let's apply this to the quarter that just ended. Let's go through the math. The actual global average alumina index decreased by -- pardon me -- $44 a metric ton Q2 versus Q1. In Q1, it was $340 a metric ton, the prompt price average daily -- pardon me. And in Q2, it was $296. So a $44 million (sic) [$44] decrease.

However, on a 3-month lag basis, the average price increased. It was $308 in the first quarter per metric ton on a 3-month lag basis and $340 in the second quarter.

Now let's do the math to get the costs. As those of you who follow the industry know, we use just shy of 2 metric tonnes of alumina per metric ton of metal that we produce. So as a proxy for our production, what you could do is take our Q1 global sales, those were about 186,000 metric tonnes, and multiplying this out, 186,000 metric tonnes times 2 to get how many tonnes of alumina we would have needed. Times that $32 increase based on the lag price. You would have got a $12 million increase in our cost of sales for alumina Q2 versus Q1.

Shelly will give you the detail in a couple minutes, but the actual cost was up $14 million versus that $12 million, which tells you that the effective lag was somewhere between 2.5 and 3 months. This kind of anomaly only occurs in periods where the alumina index is very volatile. And as you know, it's been quite volatile in 2017.

If you apply this same logic, you can predict that our alumina cost will decrease by $15 million, is what the math would tell you, Q3 versus Q2. That's a number that you would've gotten if you had used the prompt price to try to predict the Q2 cost. And then again Shelly will go through the other factors that were out there -- that were at play during the quarter. Let me move on now.

As you'll recall, we filed an antitrust lawsuit in South Carolina several months ago. Our suit claims the power provider is an illegal monopoly under federal and state laws. As a reminder, further, we've never sought any subsidy of any type from the power company or from the state or from anyone else. We must however, be allowed to buy the remaining 25% of our power needs from the competitive market in order to make the plant competitive, which means 100% of the power required for the entire plant to run -- as you know, it's still running at 50% of capacity. We continue to have confidence in the validity of our suit, and we await the court's decision on the power company's motion to dismiss the case. And we continue to be ready, importantly, to restart the second line at Mt. Holly if we can achieve full market access.

Moving along, Pete in a minute will take you through some detail on the macro environment, but at a very high-level, the industry fundamentals have remained reasonably consistent with what they were when we last spoke with you. Demand in our markets remains stable at decent levels. Value-added product premiums have stayed well below 2016 levels as we predicted, but haven't deteriorated any further from the beginning of the year.

On the supply side, we've seen no real change to what we reported several months ago. We have begun to see some modest curtailments in new product stoppage in China, but the monthly export reports continue to print records regrettably. So we will obviously be watching this carefully over the month to come.

That's a segue into a brief update on the fair trade front. We had an important development just a few -- pardon me -- days after we reported first quarter financial results. The President signed an Executive Order under Section 232. In that, he ordered the Commerce Department to study the implications for the national security of aluminum imports.

Some of you may have noted there was a hearing at the Commerce Department in June at which we and many others testified. And obviously there has been a lot of discussion since then in the media and otherwise.

In our opinion, it's difficult to predict at this point where this process heads and with what timing. But the order itself and the ensuing dialogue, in our strong opinion, is an indication that the administration understands the problem, appreciates the gravity of it and the requirement for quick action. So obviously, we'll be staying tuned and watching this space very carefully. And with that, I will hand you over to Pete.

Thanks, Mike. If you move on to Slide 5, please, I will take you through the current state of the global aluminum market. The cash LME price averaged $1,911 per ton in Q2, which reflects a 3% increase over Q1. In the last month, aluminum prices have climbed as high as $1,925 per ton, and are currently sitting right around $1,900.

Delivery premiums in both the U.S. and Europe have seen some pressure downward in Q2. Regional premiums averaged approximately 9 cents per pound in the U.S., and approximately $144 per ton in Europe. Currently, premiums are approximately 7.5 cents per pound in the U.S. and $140 per ton in Europe.

For the first 6 months of 2017, the global aluminum market recorded a surplus of 150,000 tonnes. This first half surplus is due to increased Chinese production, which increased 16% year-over-year.

In the second quarter of 2017, global aluminum demand grew at a rate of close to 6% as compared to the year ago quarter. We saw 8% year-over-year demand growth from China and 1% in North America.

Year-over-year global production growth was up 9.2% in Q2, driven almost entirely by increases in Chinese production.

Okay. Moving on to fair trade. We have spoken in detail about the need for immediate action to be taken to stop illegal Chinese excess production. Efforts to create a global forum or other multi-lateral discussions to reduce overproduction have not succeeded. However, since the U.S. has taken specific action of filing the WTO trade case and launching the Section 232 investigation, we have begun to see some initial positive responses from China.

First, the Central Chinese Government, along with several provinces, announced intended curtailments to occur during the upcoming winter heating season. This has since been followed by another announcement from the central government proclaiming the curtailment of illegal, unpermitted expansion projects in several provinces. Initial reports out of China suggest that some of these cuts are indeed occurring. If these policies were to be fully implemented, experts predict the market could potentially see millions of tons per year of production likely to close this year.

While implementation of these policies remain uncertain, we believe that Chinese overproduction will only be addressed in response to concrete actions with clear remedies. Therefore, it is imperative that the U.S. Government follow through on its Section 232 investigation with actual remedies to stop illegal global subsidies and overproduction. Other affected regions should also take action, whether through pursuing the current WTO case or otherwise.

This problem will only be addressed with continued vigilance. And we are confident that the U.S. and other affected governments understand this. With that I'll turn the call back over to Mike.

Thanks, Pete. If we could move on to Slide 6 please, a couple comments on the operations for the quarter. As you can see, we had a generally good quarter in all of the operations. First on safety, you can see we booked generally decent progress. Hawesville started off the quarter reasonably poorly, but the environment has been very good since then. And as you can see the other plants maintained a reasonable performance.

As we've said, this is an area where one really needs to press for constant improvement or it's very easy to fall backwards. We had really good performance over the last couple months, especially in Kentucky, in the middle of the summer heat season. We put a lot of appropriate focus on preventing heat stress injuries, and that effort is paying off significantly.

Moving down you see production is stable during the quarter. Shelly will detail it for you. You'll see that sales volumes are actually down a little bit and this is simply due to timing.

Moving along as you can see production efficiencies, again, show a stable set of plants. Hawesville especially is performing really well. The plant had a very good first quarter and has only improved in Q2 and into the summer. We're really proud of the team there and importantly, this kind of performance will allow us to make a more confident decision to restart capacity when conditions warrant.

Lastly, on conversion costs, again, generally in line. All of our plants are facing rising carbon costs, and we expect this trend to continue through the rest of the year, and at this point as far as we can see into 2018. As you can see, good performance at Hawesville and Sebree offset that impact. And at Grundartangi that increase that you see in addition to the carbon price is all the power price, and of course, that's due simply to the higher LME. And with that, I'll turn it over to Shelly with some more detail on the financials.

Thanks, Mike. Let's turn to Slide 7. I can walk you through some high-level points on the second quarter results. On a consolidated basis, global sales were up 6% quarter-over-quarter, reflecting favorable market pricing. This is partially offset by lower shipments in Q2 due to timing of delivery. Both U.S. and Icelandic realized prices were up about 8.5% during the quarter. And this is in line with the improvement in the delivered aluminum price on a 2-month lag. Value-added product premiums in the U.S. remained depressed in Q2, and we expect it to continue well into 2017.

Looking at operating results. Adjusted EBITDA was $34 million this quarter, which is up $12 million compared to Q1. And we had adjusted net income of $0.01 per share. Adjusting items this quarter primarily related to the mark-to-market of our aluminum forward sales contract.

Our Q2 results reflect the impact of the significant strengthening in the Icelandic krona during the quarter, including a $1.2 million increase in cost of goods sold and another $1.8 million charge in other expense.

We also had a $3.5 million realized loss related to the cash settlement of our LME hedges that matured during the quarter. As a reminder, we have forward sales contracts for approximately 16,500 tonnes per quarter at $1,700 per ton through the end of this year. Each quarter we net settle in cash the difference between the market price and the hedge price multiplied by the volume. So if the LME price remains above $1,700 per ton in Q3 and Q4, we will record additional realized hedge losses in those periods.

Okay, just to be clear, the currency impact and realized hedge loss were not included as adjusting items. So both reported and adjusted net income included the $6.5 million negative impact from these 2 items.

Turning to liquidity. Cash increased by $5 million to $131 million as of June 30, and availability under our revolving credit facilities increased by $36 million to a total of $136 million. Most of the increase in revolver availability was driven by the $25 million [LC] reductions that we talked about on our last call, along with an increase in our U.S. borrowing base.

Okay. Let's turn to Slide 8 and I can walk you through our Q2 bridge of adjusted EBITDA. As I mentioned earlier, we saw a nice increase in transaction prices for both our U.S. and Icelandic operations. This improvement in pricing increased EBITDA by $30 million in Q2. Going the other way, we had increased alumina cost of $14 million, reflecting the 3-month lag in realized alumina prices. As Mike mentioned, we expect to see this increase reverse in Q3, based on lower alumina index prices over the last several months.

Other raw material costs were up about $1 million during the quarter, primarily due to higher carbon anode costs. Coke and pitch prices, which are the key raw materials for carbon anode production, have continued to strengthen. And we expect to see further impact from higher carbon costs in Q3. Our current estimates indicate that this could be around $5 million in increased costs next quarter.

Our power cost in Iceland and in the U.S. were each up by about $1 million quarter-over-quarter. In Iceland, the increase was driven by higher aluminum prices. And for U.S. plants the increase was a result of slightly higher power and natural gas prices. As discussed in our last call, we had a significant decrease in our power capacity costs for Hawesville and Sebree, which started in June, but the full effect of this reduction won't be realized until Q3.

Okay. Let's turn to Slide 9. We'll take a quick look at cash flow. Capital expenditures during the quarter were $10 million, bringing us to a total of $19 million year-to-date. For full year 2017, we expect CapEx to be around $30 million.

We also made our semi-annual interest payment of just under $10 million in June, and had a $10 million increase in working capital during the quarter. The build in working capital was primarily due to increased inventory at Mt. Holly, which we expect to recover in the second half of the year, and timing of receivables collection. And with that, we can open up for Q&A.

(Operator Instructions) And our first question will come from David Gagliano with BMO Capital Markets.

I just have, first of all, just a couple -- or a few clarification questions from the prepared remarks. I tried to write these down as quickly as I could, Shelly. But first of all, I wonder if you could just follow up again and clarify. First of all, the $6.5 million hit to the adjusted EBITDA in the second quarter -- that's in adjusted EBITDA, sorry, for currency and realized losses on hedges, what did you say the expected impact was going to be for 3Q and 4Q? I missed that part.

So I didn't give the specific impact for Q3 and Q4. What I did is I walked you through the math. So we've got 16,500 tonnes per quarter of hedges locked in at $1,700 per ton. You multiply that out, the difference between your assumed market price and the hedge price, times the volume, that will give you the realized price that we would expect to settle in Q3 and Q4. So use your own LME price for Q3 and Q4 and you can come up with that. Currency, obviously, that's going to move around intra-quarter. So we did not put out an estimate for that.

I think, David, if I could pile on just presumptuously. I think Shelly hit it. The best to me anyway, the simplest way to model it is on the revenue side is just from the bottom-up rather than the tops down. So as she said, use your estimate as to what the quarterly sales are going to be in tonnes and assume that all but 16,500 of the tonnes get the full market price, i.e., LME plus Midwest on a lag basis as you know. And then for 16,500 of those tonnes, obviously, you get the whole Midwest, but you'll only get $1,700 in LME. And that to me instead of doing the full math and then subtracting what the hedge realized settlement would be, that to me that's the simplest way to do it.

Okay. That's helpful. Are you continuing to hedge into 2018 at the same pace?

No. No. We haven't put those hedges on. I mean, look, this is the perennial wonder of hedging, where you've seen -- I don't know, I wouldn't say never get it right, but it's always -- this is always a big question. So we put these on to protect Mt. Holly around the time where, let's see, we would have booked these, Shelly, in -- spring of '16. We're all heading around the same place. Where we saw the market shoot up pretty nicely and we could achieve $1,700 for the calendar '17, and it was a price where with the Midwest premium we thought -- and it is, Mt. Holly is cash flow good. And of course, now we're looking at some opportunity costs there. But that's a long winded answer. The short winded answer to your question is, no, we haven't sold a tonne since then.

Okay. All right. And then just on the Hawesville power cost. I think Shelly, you said there is the full impact of the hit will be in Q3. I think is what you said. And I was wondering if you could quantify the delta between Q2 and Q3?

Yes, so it's actually a benefit, Dave. So as we talked about on the last call the change in capacity cost, it translates to an annual EBITDA benefit of about $15 million per year.

Between Hawesville and Sebree combined.

Combined, the total impact on the annual EBITDA, it was only implemented June 1 of this year, so we had 1 month reflected in Q2. In Q3 we'll have a full 3 months of that benefit.

Okay. And then just last one, bigger picture question. And I think you touched on this last quarter, but again, if you can touch on it again this quarter. What would it take to restart the capacity? Can you give us some more concrete numbers in terms of pricing? And for how long?

Sure. Yes. It's pretty simple -- David, you're talking specifically about Hawesville?

Anything and everything you are willing to volunteer.

Yes, look, I mean, when you said pricing, I mean, pricing is only relevant -- metal pricing to Hawesville. I mean, at Mt. Holly, the answer is today, as soon as we can get that last 25% of our power from the market. And so that -- given Mt. Holly is very attractive cost structure, productivity of the plant, the newness "relative sense of the technology", if we could achieve through our litigation or otherwise, we hope the latter, the last 25% breakthrough to get to market -- as you know, we're buying 75% today -- we'd restart that today. It would take several months obviously to recruit people and to do the necessary -- some necessary deferred maintenance and then rebuild some pots, and then of course, start up the pots on an orderly basis. But we'd literally start on that process today. At Hawesville, I mean, right now, the plant makes money. And current basis, costs and revenue of commodities, I would say if we had some confidence that sort of this environment or maybe a little bit better, had some legs, we would be nearing that decision or at that decision. So that tells you in a long winded sense, current prices -- if we had confidence that the current price environment at the very least was going to, David, to persist. We're just not quite there yet, because we haven't seen any -- as Pete detailed on the fundamental side, we haven't seen the excess supply come out of the market yet.

Our next question is from Brett Levy with Loop Capital.

Can you guys -- can you talk about 232? I mean, I know Wilbur Ross was in The Wall Street Journal kind of saying, I mean, obviously, Trump has got the health care, taxes, immigration, God knows [there's many things], stopping leaks, firing and rehiring, a bunch of things on his agenda. In terms of 232, where do you think it is as a priority, I mean, I think, the other thing that sort of came up is that the run-up from October, like, [458 to 620 hot band] has somewhat impacted this yield urgency of 232, even though aluminum is up, it's not up like that. Is there any way of unbundling aluminum from steel in 232, or are they both kind of wed at the hip?

Thanks, Brett. I mean, I'll give you some complete and utter rank speculation, because that's all we can give you. On the latter we've always said, again, this is just speculation on our part, but we think pragmatic speculation. We've always said that we thought, and I think, a lot of other people did, that said that ally was somewhat behind steel just because the President put out the steel order first. I think it was about a month before, give or take, maybe 3 weeks, maybe 5 weeks, but let's call it a month. And so and they've clearly been working hard on both, but they're working really hard on steel. And the hearings in front of ways and means and all those briefings and all the rest have to our knowledge anyway focused on steel, have been about steel. So I mean, that's the -- I think, that's at least a stab at the -- your second part of your question. Unbundling, I don't know, whether they're bundled or not. They're obviously 2 very different markets with very different -- extremely different dynamics, very, very different dynamics. On the first, Brett, it's going to sound like we're dodging you, but I'm going to quick kick here. I'm not even going to let it go to fourth down. It's just, this one is really hard. As you say, there is a lot of factors at play. There's discussions with China, there's political issues going on. But the good news is that based on obviously things like Secretary Ross's op ed and in all the processes that we've seen and dialogue that we were having with the administration and those others in the industry, we're convinced that they are focused. I mean, they are -- this one hasn't been shunted to the side. This one hasn't been put out to pasture. They're working actively on it. And it's just hard to speculate sort of in what form it comes and when it comes. It really is. It would just be guessing.

And the second one, then I'll go back in queue. Is that in terms of the costs and your outlook for the costs. How much higher are carbon costs at this point and going forward? And then with respect to alumina specifically, is there a way or a need at this point in your view to lock in costs or secure an agreement? I mean, I know obviously there's been a historic kind of [gramercy] that you make a link to a lot of your -- of all your raw materials. And the guys who own that are going through their own challenges and maybe they cut you a deal in some way. If you look at specifically carbon costs and your outlook for alumina costs, are you looking to kind of lock-in or come down at this point?

Yes, it's a great question. That's really hard in both of those markets. I mean, take alumina first, I take your point on how somebody who owns capacity, whether it's by owning an asset or has capacity locked up is long alumina, fancy words. But remember, as you know, Brett, these are -- there's no forward market in either of these. And so if you're an owner of -- if you're long let's say alumina, you can always sell at the index price. Right? The next spot cargo is going to go at the index price. And then it's just a freight differential. It's a global commodity as we've always said, no great wisdom in that statement. And so that's a long winded way of saying, doing what you're suggesting would be really tough, other than -- short of, I should say, maybe you're suggesting actually owning, investing in alumina production capacity. But short of that, there's really been no, to our knowledge, even attempt at creating any kind of organized -- certainly organized forward market in ore, in alumina. And regrettably, the same is true of coke. There's a lot of factors at work in the coke market, that price sort of trades marginally off of the China price, and that's what you're going to get in Europe and that's what you're going to get in Australasia and that's what you're going to get in the U.S. And that price has crept up over the 6 months by, depending on the grade, sulfur content, $50 to $80 a metric tonne. As you know it takes, let's call it, 55 -- 0.55 of coke -- I'm talking about coke, of course -- to produce a ton of metal so -- in most plants. Call it between 0.5 and 0.55. So you kind of regrettably, unless you own coke capacity, which not even many of the integrated prime producers do, this is one you're kind of exposed to regrettably. We saw it nicely in the first half of this year compared to the second half of last year. And now, it's coming right back to, quite frankly, Shelly and Pete, where it was 12 months ago. Yes.

We'll go next to Novid Rassouli with Cowen & Company.

Just to nail down on costs a little bit more. So you guys are expecting coke and pitch prices to increase further. I think you said $5 million, but you'd see cost savings of alumina of $15 million. So basically that $15 million bar that we saw in the bridge would be a benefit of $10 million in the third quarter roughly?

All else being equal, yes, good add. You got it.

Okay. And then just as far as alumina, coke and pitch, it sounds like you expect coke and pitch to move higher in price and maybe alumina -- if you could just -- maybe just give your just quick high-level, what you think is driving these and in which direction through the end of the year?

Sure. Coke and pitch, I guess, if we -- gun to your head, is probably a risk on the upside, meaning bears for us and for all users. While there's been some significant price increases put through and that -- those are embedded in that estimate that Shelly gave you up to $5 million, that's all 4 of our plants combined. Could it continue to run? It's hard to say. A lot of it is based on, as I said, of the balance in China and what occurs there in terms of coke plants ultimately coming down. On alumina, our view is, I think, Pete may have said it, we think it's reasonably supported, right, Pete, around the current price $300 plus or minus is kind of when you look at where the global cost curve is and where you sort of the third runs into the fourth quartile on the cash cost curve, $300 seems like a rational, if that's the right word to use, price. Would you add anything to that?

No.

Okay, great. That's perfect. And then just one follow-up. I think last time you had mentioned that you had some smelter expansion projects in China that were to be suspended. I think it was about 2 million metric tons of capacity. I just wanted to see if there is any new developments in China as far as suspensions to projects or capacity, just kind of giving us more of an inside view of kind of what's taking place there and the validity of kind of this view of restructuring the steel and then same thought process for aluminum. Just any sort of data points would be helpful.

Thank you, great question. And you hit -- in my opinion, any way you used the exact right word, validity. I couldn't think of a better word to use. And so look, Pete -- you have to put them into a couple chunks. Pete referenced the heating degree days, closures, those are seasonal and we think those will come off and that's upwards of, what, Pete, 6 to -- on an annual tonnage basis.

About 30 percent of...

So hundreds of -- over [1 million tonnes] on an annualized basis coming out on an annualized basis. Yes.

Yes. So if you're just talking about during -- how much will come offline, it's a much bigger number, but it's not all year round.

Correct. Yes. Pardon me. I think we're all saying the same thing. On the rest, you referenced the right number. We've seen estimates as these are added up as low as 1 million, 1.5 million tonnes and as high as 6 million tonnes. And people can point with precision to how that 6 million tonnes are made up because what they've done, we've gone through this before, is they've simply taken the pronouncements from the central government at their face value and calculate, added it up those plants and expansions as Pete correctly said that were built without any permits -- environmental permits, construction permits, operating permits, any of those 3. And you can get up easily to 6 million tonnes, easily. With 1 producer, you can get up to several million tonnes. I think everybody's familiar with who that producer is. So it's one -- it's a great question, 1.5 million tonnes seems to be for what a consensus is worth sort of the low end, but taking the pronouncements at face value, you can get up to 4x that pretty quick.

Our next question is from John Tumazos with John Tumazos Research.

Forgive me if I just naïvely read the International Aluminum Institute statistics. And if you look at the alumina output in relation to the ingot, at a 1.94 to 1 ratio, which used to be okay, maybe it's a couple points lower now.

Yes. It is.

Giving credence to that adjustment, from February to May, there was 7% too much alumina output in relation to the ingot, all in China where the alumina output is reported up 29% for the first half. As though they were getting ready for 10,000 ton a day more ingot. Then in June the ingot went up almost as though the -- fulfilling the forecast of the high alumina output in China and the excess was only 2.6% of June alumina in excess of the ingot. You may recall from April 2014 after the Indonesian embargo until about September last year, there appeared to be a shortfall in bauxite and alumina potentially. Do you think the devil could be fair in advocating that the Chinese are just ramping up? And maybe they're overproducing a little bit in the first half of the year or the first 10 months of the year to prep for the winter and they'll cut from -- in the winter from just -- from a very high base?

John, I think I followed your math. But my problem with what you -- the answer is, I don't know. And here's where the word that you used on which I'd key. You said they, and I don't think there is a they here. Your they, at least to me, implies organized behavior like, group, coordinated organized behavior. But based on what we've seen and I think others in the industry would concur is just, I mean, I think the fact alone that millions and millions and millions and millions of tons got built unpermitted is some evidence that there is no they. That you have lots of market actors doing their own thing to use simple terms and simple words. And so your thesis, I think I understand it, I will let Shelly and Pete comment. It may well be true, but I don’t think the data necessarily like track organized purposeful behavior behind them. It may turn out to be true, but I don’t -- Shelly?

I agree. What you are saying makes sense.

I just I wish, I wish, we wish there were more, sort of, controlled -- maybe that's not the right word -- behavior. Maybe there would be a different market balance right now.

That might have just been my poor semantics, that whoever is producing the alumina must think there is a smelter that's going to take it.

Sure. Fair. Fair.

(Operator Instructions) And we'll go next to [A. J. Lilly with Southpaw].

I was just wondering with regard to alumina. I note that Alcoa are thinking of starting another smelter. And in terms of the price you get, do you have to pay a premium like a regional premium on the alumina cost because there might be a certain amount of capacity in the North American region and now multiple smelters want the alumina from Gramercy. That was my first question. The second question was, second question was around green aluminum? And what you think that impact is going to have in North America broadly? I heard Rio talking about that. And then thirdly -- my third question was just around 232 and like whether you think there could be any perverse outcomes where recycled capacity of aluminum in North America, start to gain market share and take advantage of that versus primary and if there was a reduction in the amount of imported material?

Thank you very much. First, just the facts on alumina. So there is a slight when I'd call regional premium impact, but nothing near metal. There is a global price and then there is for those of us who buy or sell in the so-called Atlantic region, there is an Atlantic premium or discount generally the Atlantic is traded at a discount. The Atlantic has been better supplied than the average of the rest of the world and that's the case today. And I can't speak with certainty and I really don't like to talk about peers in the industry. But our supposition is that Alcoa of course has alumina in their own system, so they won't be out in the merchant market seeking to procure. The other just from a factual standpoint, I think, as you know, we've disclosed this in the past. We have a multiyear agreement with the Gramercy refinery for a significant portion of its output. So at least for the next several years -- you asked a decent question, and of course, if there were other restarts, it could be a factor, but for the foreseeable future, I wouldn't see this as an issue. And then Shelly notes correctly -- do you want to just [do this]?

Yes. Sure. And there is a refiner actually in the Atlantic area that would supply the Atlantic area in Jamaica that's anticipated to come on later this year so that should add incremental alumina supply for our region.

Yes. From what we understand, they are actively working. They've restarted. They've had some production issues as is typical for a larger refinery restarting, but they intend to run it. They bought it and they paid value for it and they've rehabbed it and it's a Chinese owner. So we would expect it to run. And as Shelly said that will significantly increase the oversupply in the Atlantic basis to a -- basin, pardon me, to our benefit. Second part of your question on green ally, that's a great question. I would say that whole process is at its early stages in the U.S. In Europe, we're a significant market participant in Europe through our plant in Iceland as you probably know. And we're participant in the European Aluminum Association. There's been a lot of work and talk. And there is a bit more momentum there. The ultimate question is the obvious one, whether consumers ultimately are willing to pay a premium for it. I mean that's what interests us first and foremost besides being a good citizen and all the rest, which, of course, our production in Iceland has no carbon footprint. Third, that's a good question on 232. Obviously, scrap is dependent on a lot of things including spreads and the end use for the product, some of which can take high scrap input like can sheets. Some like our high-value billet, high purity, of course, and others need to be made purely from prime. So we don't -- that's not a factor that we see. Unintended consequences is a good term, but this is not one that we're concerned about.

And we have no further questions in queue. Do you have any closing remarks?

No, other than to thank you all for joining us and for a good call. And we look forward to speaking with you in a couple months. Take care.

Thank you, then ladies and gentlemen, that does conclude our conference for today. Thank you for your participation and for using AT&T Executive Teleconference. You may now disconnect.