Century Aluminum Co (CENX) 2016 Q3 法說會逐字稿

Ladies and gentlemen, good afternoon. Thank you for standing by, and welcome to the Century Aluminum Company third-quarter 2016 earnings conference call. (Operator Instructions). As a reminder, today's conference is being recorded.

I would now like to turn the conference over to our host, Mr. Peter Trpkovski. Please go ahead.

Thank you very much, Tom. Good afternoon, everyone, and welcome to the conference call. Today's presentation is available on our website, www.centuryaluminum.com. We use our website as a means of disclosing material information about the Company and for complying with Regulation FD.

I would like to remind you that today's discussion will contain forward-looking statements related to future events and expectations, including our expected future financial performance, results of operations and financial condition. These forward-looking statements involve important known and unknown risks and uncertainties which could cause our actual results to differ materially from those expressed in our forward-looking statements. Please review the forward-looking statement disclosure in today's slides and press release for a full discussion of these risks and uncertainties.

In addition, we have included some non-GAAP financial measures in our discussion. Reconciliations to the most comparable GAAP financial measures can be found in the appendix to today's presentation and on our website.

With that, I would now like to introduce Mike Bless, Century's President and Chief Executive Officer.

Thanks a lot, Pete, and thanks to everybody for joining us this afternoon. If we could turn to slide 4, please, just give you a quick rundown on what has been going on here over the last couple of months.

I am going to ask Shelly to give you some detail on the market environment in just a couple of minutes, but let me just make a couple of quick comments to put the quarter into context.

It goes without saying we are seeing mixed signals out there. This is one reason we believe why the LME price appears to be trading within a range. The most significant factor by far continues to be the excess supply in China.

We began the year by seeing a relatively modest volume of capacity restarts. But this pace, as expected, has picked up, and we expect to see over 1.5 million tonnes of restarts year-to-date and a little bit more trickling through. The market's expectations for new capacity starts now appears to be coming true, with expectations for over 2 million metric tonnes of new capacity starts by the end of this year.

This continues to puzzle us, as most factors are not supportive of these capacity starts. We are seeing rising coal and energy prices in China as well as, most recently, rising alumina prices, which are a bit of a Catch-22 as far as we're concerned, as they are driven primarily by unnecessary increases in production of aluminum in China full circle.

The demand picture in China has been a bit better than originally expected, with government stimulus spending on large projects driving demand and a spike in property values in many regions driving residential construction, as you probably read. It is clear to us and most others that these trends simply can't continue over the long-term.

Otherwise, in our markets we are seeing mixed trends. On the one hand, you have seen delivery premiums growing in both the US and the EU. On the other hand, value-added product premiums are looking weaker to us currently and going into 2017, largely driven by imports from regions flooded by materials coming out of China.

Some markets, however, do remain attractive. For example, the EU foundry alloy market looks good to us, and that's a very good business for our plant in Iceland, Grundartangi.

Turning to operational and financial performance during the quarter, Rick is going to give you more detail in just a couple of moments, but just a couple of quick comments. As you have seen if you have been able to read through the press release quickly, the results for the quarter were impacted by higher power prices in US and by rising alumina prices.

If you have had a chance to take a look, you have seen that adjusted EBITDA was down $26 million quarter to quarter. That's from Q2 to Q3, obviously. The impact of those two factors, higher US power prices and alumina prices in aggregate, drove more than 100% of that change.

The power price increase came fast and was very steep due to the very hot summer weather in the United States. That having been said, delivered power prices to our plants are now down generally to pre-summer levels, obviously a good trend to see.

The alumina price has seen a rapid increase over the last couple of weeks. We believe that the startups in China create somewhat of a herd mentality, and we expect this trend to moderate. There have been no substantive changes in the alumina markets, again, other than the restart of Chinese metal capacity. And of course we're watching these trends closely.

We did have some minor operational issues at Sebree and Grundartangi during the summer. These plants usually deliver rock-solid performance. These issues are now behind us, although they did cost us some volume in Q3. And I will give some more detail on that in just a couple of moments.

Moving down, those of you who have been following the Company for some time will remember that in 2009 we terminated the retiree medical benefits for our folks at Ravenswood. The liability before that termination had been well over $100 million. The related litigation has been winding its way through the legal system now since that time.

The risk to us, obviously, was ultimately that it would go to trial in a West Virginia courtroom, with the obvious unpredictability of that outcome. We have now reached a tentative agreement. It is subject to a definitive agreement of the various parties and approval of the court. We believe the settlement is a very good mitigation of the litigation risk to the Company. It also does provide some support to those folks who worked very faithfully for this Company, some for decades and decades.

The total value of the settlement is $23 million. This is the future value; it is paid over 10 years. And there is no risk -- there is no variability of that amount to the Company. Our obligation is to make a fixed payment one time per year.

Lastly, we have seen a lot of activity on our fair trade efforts over the last couple months. And especially over the last month or so and even couple of weeks, we have seen a lot of momentum building, to give you a couple of examples.

You may have noted that the Aluminum Extruders Council joined our task force some time ago. This is important to us, as it demonstrates the broad-based support amongst to the industry for these efforts.

You may have seen, also, this was widely reported that last Friday eight United States senators sent a letter to the US trade rep urging the immediate filing of a WTO case, and this is the specific action obviously that we have been recommending for the last several months -- several quarters, to be precise.

In addition, we have seen additional support from other members of Congress over the last several weeks. Last week, a statement came out from the EU specifically relating to subsidized excess aluminum production in China. And over the last couple of days at the WTO itself, several member states have requested that this specific issue be investigated.

This effort has been quite significant, as we have told you in the past. Just to give you an example, our researchers in China have now translated over 100,000 pages of information. We continue to believe without condition that the case is absolutely irrefutable, that the illegal subsidization by the government of China of this industry represents a textbook case of the violation of WTO rules, and we are waiting for definitive action by the government.

And with that, I will turn it over to Shelly.

Thanks, Mike. Move along to slide 5, please. I will provide some comments here on the industry environment.

The cash LME price averaged $1,620 per tonne in Q3, which reflects a 3% increase over Q2. Prices have continued to be quite volatile, trading as high as $1,690 and as low as $1,550 over the last three months. Today, the cash LME price is close to the high end of the range at $1,682 per tonne.

The downward trend we saw for regional premiums over the last many quarters appears to have reversed. Delivery premiums in both the US and Europe have strengthened recently, and continued upward pressure on regional premiums is expected for 2017.

The US Midwest premium averaged $0.064 per pound in Q3 but is currently sitting just [about] $0.07 per pound. In Europe, the duty-paid premium averaged $120 per tonne in Q3 and has since traded up to $135 per tonne.

The global aluminum market was in a deficit again during Q3, with a shortfall of about 300,000 tonnes. We continue to see good demand growth with a 4.7% increase in global consumption year over year.

Chinese demand growth at 6.7% was better than anticipated, with continued strength from the construction sector. But as Mike mentioned, we continue to believe this growth level in China is not sustainable, as current demand is being propped up by government stimulus and other temporary factors.

Global primary aluminum production was up 3.1% in the third quarter, driven by startups and restarts in China. We expect to see meaningful capacity additions from China in Q4 as growth projects continue to come online and previously curtailed capacity returns to operation despite rising costs for key raw materials. These additions to production will continue to outpace consumption, further increasing China's already oversupplied position.

Most industry experts are now anticipating that 2016 will see a modest deficit in the global aluminum balance, with a sizable deficit in the Western world being largely eroded by the ongoing surplus in China.

Okay, just a couple quick comments on the alumina market before I hand it back over to Mike. Alumina prices have been increasing recently from $235 per tonne when we last spoke with you in July to their current level close to $290 per tonne. As Mike mentioned, the increase in price is being attributed to higher demand from Chinese smelter production growth as well as increased input costs for coal and caustic soda.

And with that, I will hand it back to Mike.

Thanks, Shelly. If we could turn to slide 6, please, a couple quick comments on the operations during the quarter. As you can see at the top of the page, safety performance was generally reasonable across the Company this quarter. We did have too many incidents at Hawesville. Most were relatively minor, but too many is too many.

We are particularly proud of the performance on safety at Sebree this quarter. During the summer we experienced very high employee attrition and somewhat -- some process instability during the height of the summer heat. These factors produced a very risky environment, and we're proud that our employees really focused on job number one: that's obviously keeping themselves and their colleagues safe.

Turning to production volume, you can see the impact of those operational issues at Sebree and, to a lesser extent, at Grundartangi. Both of these plants are expected to be up in production Q4 over Q3, and we have good confidence in that result based on the current state of the operations. As I said, both plants have returned to full stability very quickly.

Turning now to production metrics, as you see, Hawesville -- the operations at Hawesville have remained very strong. Just to give you a sense, almost 80% of the production at Hawesville is now high-purity metal. That's a record for the plant.

You see the impact of the summer potline instability at Sebree there with current efficiency, power and raw materials efficiencies all down Q2 versus Q3.

Turning to conversion costs again, just to note, conversion costs here are all cost other than alumina, of course. And these, again, driven by US power prices. Let me just give you some details so you can see that.

At Hawesville quarter to quarter -- again, this is all Q3 over Q2 -- power prices were up 35%. Those of you who follow the MISO markets of Indiana Hub price already know this. Power up 35% Q3 over Q2. Labor and other controllable costs were down 11%. Sebree power up 29% Q to Q. Controllable costs up a little bit, as I said, due to the instability. Mt. Holly power up 12%. Controllable costs down 2%.

So, you can see the operations in general turned in pretty good performance on costs that they were able to control.

In addition, as I said, the power price drag appears largely behind us. Power prices at our Kentucky plants have come down nicely. We are almost past the utility maintenance season. And natural gas prices, for those of you who follow these markets, have finally now begun to come down nicely over the last couple days; obviously watching that closely.

And we're also watching the alumina price closely. As Shelly said, it has had a very rapid and quick run-up. We continue to be convinced it is based on people chasing the price as it goes up in anticipation of restarts in China, and we do expect it to moderate.

Mostly, we are pleased with the status of the operations. Value-added production is now a significant portion of our product portfolio. As a reminder, we produce billet, high purity metal, foundry alloy and molten aluminum. In aggregate, those products represent about 85% of our US production and about 25% in Iceland.

Importantly, we have the flexibility to flip quickly between these product groups in order to address the market that looks the most attractive. Let me just give you a quick example. P1020, standard great, is actually look a little bit tight right now in the US specifically in the Midwest and going into 2017. That is because of a flood of imports that are chasing value-added products.

And for that reason, we have literally an almost brand-new automated sow caster that we had put in about a year or regrettably before we had to choose to close Ravenswood. And we're moving that sow caster now to Sebree to enable us to take advantage of the best market as we go into 2017 and throughout that year. And we will continue to make these kind of small investments to maintain this kind of flexibility.

And with that, I will turn you over to Rick.

Thanks, Mike. If we turn to slide 7 on the presentation, I will provide some details on the performance for the third quarter. Our net sales were up almost 2% from the second quarter on the shift from total to direct sales with the expiration of the last tolling contract and the second quarter.

On a two-month lag basis the average cash LME price was up 3% in the third quarter and the Midwest premium decreased 10%, resulting in a Midwest transaction price increase of approximately 2% quarter over quarter on a lag basis.

Realized prices in the US were down 1%, reflecting weaker value-added premium -- product premiums, as Mike discussed. We are expecting this to continue in the fourth quarter and 2017.

For Iceland, the all-in two-month lag LME and European duty-paid premium increased approximately 2% in the third quarter. Realized prices on direct shipments also increased 2% quarter over quarter, and there was an additional 10% of realized pricing attributable to the shift from total to direct sales in the quarter.

On a consolidated basis, global shipments were down 2% in the third quarter of 2016. Production levels were essentially flat quarter over quarter.

Turning our attention to operating profit, we are reporting an adjusted EBITDA loss this quarter of $5 million, a decrease of $26 million when compared to the $21 million of adjusted EBITDA recorded in the second quarter.

The adjustments to EBITDA this quarter include costs associated with the negotiated settlement of the Ravenswood retirement medical benefits class-action suit, additional impairment of the Ravenswood assets and a non-cash adjustment to the carrying value of inventories.

As Mike discussed, we reached an agreement in principle to settle the class-action suit for $23 million. This agreement in principle remains subject to entering into a definitive agreement with the class and obtaining the court's final approval after notice to the class, which we expect in the first half of 2017.

The proposed settlement costs calls for a $5 million cash payment upon final approval and $2 million per year for the nine years thereafter. As Mike noted, we have no further obligation other than making the cash payment.

We have also reached a tentative agreement for the disposal of the Ravenswood assets, and we have adjusted the carrying value of these assets to their indicative fair value based on this agreement. We expect this transaction to be completed by early 2017.

Also, as Mike noted, the $26 million decrease in our adjusted EBITDA for the quarter is entirely driven by changes in market conditions. Let me provide a few more details of the impact during the quarter.

Higher raw material costs decreased adjusted EBITDA by approximately $15 million, led by a significant increase in the realized cost of alumina. As a reminder, there is a one- to two-month lag on alumina costs realization depending on the timing of shipments and inventory levels.

Power costs continue to increase at our Kentucky operations as high temperatures continue to affect the demand over the summer months. This resulted in a decrease in adjusted EBITDA of approximately $9 million from the second quarter of 2016. Power costs at our Mt. Holly facility were also unfavorable to the second quarter by over $2 million.

As Mike discussed, natural gas was up in the third quarter, also due to summer temperatures, and we believe it is still [elevated] on expectations for the cold winter.

Higher all-in alumina pricing, net of the impact of a rising LME on Iceland power, resulted in an increase to adjusted EBITDA by approximately $5 million. These gains were fully offset by the impact of a $5 million deduction in value-added product premiums during the quarter, as discussed earlier.

Collectively, these items resulted in an adjusted loss per share of $0.31, an increased loss of $0.26 when compared to the $0.05 adjusted loss reported in the second quarter.

Moving on to liquidity, there are no outstanding borrowings under our revolver other than letters of credit. We ended the quarter with $118 million in cash and $90 million of availability under our revolving credit facilities. Our facilities are secured by both accounts receivable and inventory and have availability under revolver -- our availability under revolver will fluctuate as our working capital levels move during the quarter.

Please turn to slide 8, and we will take a look at the cash during the quarter. Cash decreased this quarter by $11 million. Capital expenditures were $5 million during the quarter, which brings year-to-date capital spending to $13 million. We now expect spending for the year to be at the low end of our range of $20 million to $25 million provided in our 2016 items.

Cash taxes paid in the quarter were almost fully offset by improvements in working capital.

With that, I will now turn the call back to Mike.

I think, Pete, we can probably move directly to questions.

Thanks, Mike. Tom, if you can go ahead and facilitate the Q&A session, please.

(Operator Instructions). Dave Gagliano, BMO Capital Markets. David Gagliano.

I actually just had a little bit of a challenge understanding the bridge. That was just mentioned on the EBITDA from quarter to quarter. Maybe it is just easier -- can you just go through the numbers again on the driver? I got the $15 million on alumina. Can you give me the rest of the bits and pieces that explain the delta?

Sure. You got the $15 million on alumina, and then the rest of it was power costs -- $9 million from the Kentucky operations, $2 million from our Mt. Holly operations.

Okay. Then you mentioned there was a $5 million drag on value-added premiums. Is that correct, quarter to quarter?

Right, $5 million drag on value-added premiums, but you also have to look at the higher aluminum pricing. So, the $5 million from LME and Midwest transaction pricing offset by $5 million on value-added premiums.

Okay. And then did you quantify the -- roughly the EBITDA hit from the operating issues during the quarter? Not raw materials, but operating issues?

No, we didn't, David. It was a couple million dollars. It wasn't quite $5 million, but it was a couple million dollars quarter to quarter.

Okay. Then last question -- you said, obviously, alumina is priced on roughly a one- to two-month lag. Given that there's been quite a bit of movement in alumina during the course of the quarter, what was the average price that actually flowed through the cost line?

Don't have the average price that flowed through the cost on hand. Shelly, you --

The actual for Q2 was [234].

Right.

So, you are definitely -- you're going to see the prices have run up significantly from there, so there is further alumina pressure to come.

Sorry, you said the actual is Q3 or Q2?

Q3 actual, ignoring any lag, was $234 a tonne.

Got it. Okay, great. Thank you.

The key there is there is more costs to be realized -- negative costs to be realized in the fourth quarter.

Understood. Okay, great. Thanks.

Brett Levy, Loop Capital.

Tell me a little bit more about Iceland, what's going on -- Grundartangi, Helguvik, potential stuff like that. The Pirate Party.

Yes, you have been reading up. You have been reading up at the end. Let me go through it in your order rather than reverse order there.

Okay.

So, Grundartangi is kind of motoring along. And as I said, we had -- like Sebree, but less than Sebree -- some potline issues this summer caused largely by some dusting from some anode quality issues. Those -- that is completely behind us now.

So, the plant is in a reasonable steady-state at this point in time. No movement on Helguvik to talk to you about. That -- and really the reason for that in a certain respect is any movement there is going to be a result of ultimately of being able to secure power. And as I think you know, we filed -- there was arbitration filed earlier this year in this matter against HS, which is one of the two power suppliers to Helguvik.

You may remember that a similar -- in fact, frankly, the carbon copy arbitration was filed and litigated three or four years ago, with the result that the contract was found to be still in full force. HS had sought to prove and had the tribunal order that the contract was null and void.

So, we are waiting for the result, Brett, of that process, and that will be a key determinant into the next steps on Helguvik. And, as we've said, there is no firm date, but we said we expect -- we would expect given the normal timing of these things the tribunal's judgment sometime during the fourth quarter. But there is no promises. So, that's the big next inflection point on Helguvik.

You asked about the election, really, I think in your last part of your question. And speculation, just like other elections I suppose at this point, is probably pretty idle. This one will be settled even before the one in this country. The Icelanders go to the polls on Saturday, this coming Saturday, and so this will be known on Sunday. And, so, to speculate at this point in time is probably a fool's errand.

But there are lots of parties. There are more and more parties every year, and this election clearly is indicative of a healthy and robust political debate in Iceland. There are plenty of parties.

Is there a sense generally whether or not more of the parties are in favor of expanding the energy available to growth projects in aluminum or anything else or anything like that? Then the other piece of the puzzle is that, to the extent that you could hedge alumina, aluminum, anything in North America, I know historically you have tried that. Any inclination to go back to that kind of strategy in terms of the US, but not the Icelandic assets?

So, I guess it's a two-part question. Is the general tone of the election going to help Century in Iceland? And then anything you would do to mitigate what's going on in the US right now?

Sure, Brett, got it. First is -- I am going to punt. Actually, I'm going to quick-kick. I'm not even going to get to fourth down here. It is really hard to tell. As usual, there is a debate and two sides on that coin. There is significant support every time you see polling for so-called heavy industry in Iceland.

There is a chorus, obviously, on the other side. It is really difficult to tell. You're going to have to get through the election, get through the formation of a government. Of course, as I think you know, it is a parliamentary system there. So depending upon the results of the election, it could take some time -- weeks or even longer, we have seen in various parliamentary democracies around the world -- for a government to be formed just depending upon how dispersed the vote is.

So, it is hard to tell. We think we have got -- we being the industry, we have got good -- great support in Iceland, and just how that manifests is really hard to tell. What is really going to drive, again, the next steps on power for Helguvik are the contractual issues, not the political issues. It is arbitration and what comes out of that and what our next upstart steps are.

On hedging, the answer is yes. I will give it in your order. Alumina is difficult to hedge. There is a new product out there from what we understand with a relatively thin market that enables people to buy and sell forward the index price. But thus far, it is an undeveloped market.

On the other side, of course, as you are well aware, there is a really deep LME market. And the answer to your question is, yes, we have -- we are trying to be patient here and calm. But we have -- every time we feel like the price in the depth of the market has afforded, we have sold forward LME, and so I will just give you the specifics right now.

For the balance of this year and for 2017, we have sold forward 5,500 tonnes per month at a price just shy of $1,700. So, we're trying to -- there is no set bogey. But, given all the factors on any given day and again given the depth of the market and whatnot, we are, as you can tell, in the mode of trying to at least, we believe, intelligently take some risk off the table.

Thanks, Mike. I will hand it off.

(Operator Instructions). Jorge Beristain, Deutsche Bank.

Hi, Mike. It is Jorge Beristain with Deutsche Bank here. Just wanted to dive more deeply on the charges of the $27 million, I think. Can you just quantify how much of that was cash versus non-cash?

Sure, it's all right now non-cash. And so let us give you -- let me just take you through all that, because they're interconnected to a certain extent. So, the retiree medical settlement, as Rick said, is $23 million. That is a future value, a simple value, however you want to say it. It is not PV'd.

That is non-cash today, but, as Rick told you, it will be cash over the next 10 years. So it will defease itself, as you said. $5 million, assuming the court approves it sometime in the first half of 2017, and then $2 million thereafter. So, if you want to calculate a PV on that, you can use whatever discount rate you think is appropriate. But that's non-cash today, but will be cash.

The second chunk of it, the rest of it, is just a further impairment of Ravenswood, of the asset that is sitting on the balance sheet. What it is doing, Brett -- and so that isn't it. That's ultimately -- that's non-cash, period. What we are doing there is we're writing it down to the value at which we have a tentative agreement to sell it.

And, so, one way to look at this, at least in our opinion, is that if you look at the present value, again, whatever discount rate you wish to apply of that stream of payments for the retiree medical, $23 million, the cash that we think we're going to get for selling the plant and the site is in the mid-teens. It is just about equal to the PV of the settlement.

Those two weren't linked in terms of -- we didn't seek a settlement that it necessarily got us to the PV -- or to the sale price of the plant, but they just turned out that way.

Okay, but you are saying -- sorry, that what you could sell Ravenswood for is equal to the PV of the settlement, not the future value?

Correct. So, it's about $14 million give or take. And, so, that deal isn't finalized yet, but it's in its final phases of documentation.

And, so, to answer your question -- it is a long, long-winded answer to your question. Of the $27 million, $23 million will eventually turn into cash over the next 10 years. And the rest is just a further impairment charge to write down the Ravenswood asset to the cash value that we believe we're going to be paid.

Got it. Obviously, I'm a little rusty on keeping up on all the MISO power changes. But you have said that the bulk of the $27 million or more than a $27 million EBITDA hit that you took in the cost in the third quarter is abating now. So, could you just give us a sense as to how much the power costs have come down either percentage-wise or dollar-wise so we can be bracing for what 4Q could be looking like?

Yes, sure. So, as Rick said, of that $26 million delta reduction in EBITDA Q2 to Q3, about one-third, about $9 million, of it was due to the power and two -- in Kentucky, thanks, Rick. And $2 million due to the power ultimately natural gas in South Carolina, because we are floating based on natural gas there.

So, to deconstruct those two, the MISO prices have come at our plants, delivered to our plants, have come back down basically to where they were before the summer. And so if those were to continue where they are now, you should see those -- that $9 million anchor go away, i.e. Kentucky power prices, all else being equal, ought to be $9 million if they stay where they are now. $9 million lower Q4 versus Q3.

Nat gas -- I am going to have to take a guess here right now. I'm just thinking about the forward screen. It is probably show about -- you can see about halfway down. (multiple speakers). All right, we're going to say the same thing, so we will go with that.

About halfway down. So of that $2 million bump, you would only see -- you would see $1 million better in Q4. So, putting that all together, everything else equal, gas staying where it is today, MISO energy staying where it is today, Indiana Hub energy staying where it is today, as fully delivered at our plants you'd see $10 million of that bad news go away, get better in Q4 versus Q3.

Perfect. And if I could just push my luck with one more question. Could you walk us through the same alumina ladder? And is there any of the alumina that you have contracted or are going to -- are you going to be recouping in any way as the months go on? If you could be facing a tailwind, in other words, if things could improve for alumina, or should we be marking to market for what we are seeing?

That's -- let me -- go ahead, Rick.

As we talked about earlier, you should -- you will see still some of that increase that Shelly talked about during the quarter. You will see some of that in Q4. As I mentioned, you have got kind of a one- to two-month lag. So maybe by the end of the quarter you will start -- all things being equal, you will start to see some of the reductions that Shelly talked about.

Another point, Jorge, just to remember -- of the total amount of alumina we buy right now -- just let's make it simple, a little over 700,000 tonnes of finished metal production, let's just call that 1.4 million tonnes for easy math of alumina. About 40% of that we buy on a percentage LME basis through the end of this year. So, we are not chasing up or down the alumina index that you can read every day. It is a fixed percentage of the LME price.

That number hasn't been finally negotiated for 2017. It could go up or it could go down. But right now, like I said, roughly 40% of our alumina cost is pegged to the LME price, not pegged to the index price.

Got it. Thanks very much.

David Gagliano, BMO Capital Markets.

I just wanted to ask a couple of follow-ups. And I apologize; I have been jumping between conference calls here, so hopefully these haven't been covered.

Actually, one that you just mentioned, you said 40% of your alumina costs are pegged to the LME price. Can you just remind me, can you just give me the tonnes that are exposed to -- just the actual tonnes on a quarterly basis?

Sure, sure. 1.4 million tonnes is our total alumina. And somewhere around 450,000 tonnes, again, on an annualized basis for the balance of this year -- I want to stress that -- are pegged to the LME, and the rest is that we buy on an index basis. And, again, that -- David, it is under negotiation right now as to what that split will be for 2017.

And Dave, just as reminder, we put out the sensitivities at the beginning of year. So for the exposure we do add to the spot price, for every $10 change in that, annually that is $9 million in EBITDA.

That is a very good point, Shell. Thanks. So, that takes into account the amount that we buy -- obviously, index versus LME base.

Right, got it. Sorry, so you said a $10 change in the spot price is a $9 million change in the annual EBITDA. Is that what you said?

You got it.

Okay. All right. Great. Thank you. Then I just have another question. I think I heard you say that you sold forward some of the aluminum.

Yes.

And if I got the numbers right, it looks like it is about 10% of the volume sold forward, something like that?

Oh, gosh. 5,500, no, no -- oh, yes, the 5,500 tonnes a month. So for 2017 if just do the math, [66,000] divided by (inaudible), yes, maybe 8.5%, 9% just off the top of my head, David. Yes.

All right. Just conceptually, I am wondering what the thought was behind selling forward the aluminum while leaving yourself exposed to alumina?

Yes, as we said, there is no perfect philosophy -- to use a high-falutin' word -- about this. I will just say it in simple terms: we liked the forward price in terms of selling forward that reasonably modest amount. And on alumina, as I said, there is just no good liquid forward market for buying or selling alumina at this point in time. You may have been off.

There have been some markets that have opened up here, organized markets, but they're really -- if you look at the open interest, the volumes, they're really, really thin at this point in time.

So, there is not an economic ability to address alumina at this point in time.

Okay. Then just one really last question. The 40% of alumina costs that are pegged to LME through the end of this year, obviously that is under negotiation. Is the hope that that number is -- that percentage that is pegged to LME higher or lower than the 40% next year?

That's a good question. There is no right answer, I suppose, David, to that question. It depends upon one's view of the LME price. And ultimately it depends upon your ability to achieve a -- just to be blunt -- a percentage in your negotiation on the one hand or a deal around an index price on the other hand. And you have to compare those two.

I am sorry to be so --

No, I understand.

-- opaque here.

I understand. But as you negotiate now, is Century's position they want that percentage to be higher or lower at this point. Given everything that we know, would Century rather have that number be higher or lower?

David, I am sorry to evade, but it is really hard -- let me just give you an example. If you told me my LME reference price that I could buy at was going to be 12% on the one hand or 19% on the other hand, just to give you the two extremes of where the market has been since I have been in this business over the last 11 years, my answer would be really different.

Okay.

And, so, it is really hard to answer that question in that respect. We want the lowest price that we can. Now, you're going to be taking a risk on the one hand on the index; on the other hand, on what the LME that you're multiplying that percentage that you can negotiate. But it is so variable, the answer to your question, based on what the market is for LME-based contracts at the time that you are agreeing with your counterparty.

All right, got it. Thank you very much. I appreciate it.

Tony Rizzuto, Cowen and Company.

I am sorry, don't mean to beat the dead horse here, but --

Beat away.

Just on those LME-based contracts, I typically think about it -- you mentioned that range. I think about a little bit tighter range, maybe 15% to 17.5% range. And then you indicated -- so, are we to think that going into the fourth quarter and into 2017, 60% will be on -- will continue to be on the one- to two-month lag?

The convention, Tony, is increasingly whether you are buying on an index basis on the one hand or on an LME reference basis on the other hand. The convention is increasingly to have the reference price be a lag, either one or two months.

So, I would say the answer to your question either -- I will expand it. Either the 60% or the 40%, assuming that those percentages don't change and just going into 2017, I would say the answer is -- I am looking at my colleagues here -- yes.

Again, subject to negotiation, but it is increasingly the convention that metal is sold on and aluminum is bought on a lag of a month or two, or sometimes even three.

Okay. And you guys mentioned the figure there, and I didn't write it down quickly enough. But for each $10-per-tonne change, you mentioned the impact, and I didn't get that down. Could you repeat that?

Sure, Tony. It is $9 million annually on EBITDA.

$9 million, okay. And then one question, a follow-up on Ravenswood, I would assume that there is going to be no environmental or reclamation that you will be responsible for going forward.

That's correct. The price that I -- the sales proceeds that we're estimating here -- it is a net price. So, the buyer is assuming all of those liabilities, correct.

Okay. All right. Very good. I think that's all I have got for right now. Thank you.

There are no other questions queuing up at this time.

Okay. Then, we -- again, we greatly appreciate your interest and time. We look forward to talking with you in the new year if not before. Take care.

Ladies and gentlemen, that does conclude our conference for today. We thank you for your participation and using the AT&T Executive Teleconference. You may now disconnect.