Century Aluminum Co (CENX) 2020 Q2 法說會逐字稿

Ladies and gentlemen, thank you for standing by and welcome to the Century Aluminum Company Second Quarter 2020 Earnings Conference Call. (Operator Instructions)

I would now like to hand the conference over to your speaker today, Peter Trpkovski. Thank you. Please go ahead.

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

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

Turning to Slide 1, please take a moment to review the cautionary statements 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 over to Mike.

Thanks, Pete. And as usual, thanks for all of you for joining us this afternoon. We appreciate the time. If we could just flip over to Page 3 please, let me just give you a brief overview of the highlights of the last couple of months. First, a quick update on the status of our operations, obviously still in the context of the continuing health crisis. As we discussed with you a couple of months ago in early March, we essentially ceased all activity in the plants not oriented towards safe and sustainable production. That in addition allowed us increased flexibility, as we instituted a variety of measures, both inside and outside the plants to keep our folks safe.

These measures to date have produced the intended results. Thus far we've recorded a small, but very manageable, number of confirmed infections. It won't surprise you that we don't expect these policies to change in the near term, obviously given the situation in the U.S. generally. The public health environment in the Iceland and The Netherlands is a bit better. But we're still erring towards caution at each of Grundartangi and Vlissingen. Consistent with this operating discipline, we've maintained a strict management of our controllable costs. In essence, all non-required spending remains on hold. That's a change, of course, we made in March. Exceptions since the beginning has been only for spending related to safety and/or the sustainability of operations.

Recently though, we have begun to okay a few modest projects with very quick paybacks, in a matter of months type of paybacks. This operating discipline contributed to strong Q2 financial results that were consistent with our expectations. In a couple of minutes, Craig will provide more detail on the quarter that just ended. When he does that, you'll see that a drop in the LME price plus a drop in regional premiums, principally the Midwest premium of course, reduced EBITDA by $32 million from Q1. So that's LME and premiums combined $32 million Q1 to Q2 down. That's just a couple million dollars worse than we forecast to you in late April, as the last month of the quarter, as you'll recall, remained unpriced at that time, as it always does. And as you'll also recall, May prices, LME prices remained very low in Midwest.

We were able to partially offset this reduction via a [$313] million decrease in controllable costs and raw material prices. In addition, cash flow was quite strong. Operating cash flow for the quarter was $37 million and CapEx only $4 million. Cash on hand increased by $28 million. Craig will also walk you through the expected changes in realized commodity prices, Q2 to Q3, obviously the quarter that we're now in, and the impact that that change will have on our reported financial results in Q3. Obviously, the impact of those very low metal prices in April and May will be felt in the Q3 results, due to our normal 2-to-3-month lag. Those prices appear now to be behind us. At the current commodity prices, the company's performance is materially better than what the Q3 reported results will look like, and Craig will give you more detail and some data on all of that.

Moving along, as I'm sure you saw, we refinanced the debt issue that was due to mature in June of next year. That transaction only closed on the 1st of July. So obviously it's not reflected on the quarter-end balance sheet. Obviously, there's a higher coupon in the new issue, but we believe it was the right thing to do to get this done at this time. The terms of the new notes are attractive for early redemption, assuming conditions continue to improve.

In just a minute, Pete will provide you our outlook on the sector, including global supply and demand. I'm sure you've noted recent data indicating a decent upturn in global manufacturing activity generally. You all follow the macro data, so I'm not going to spend your time now providing further commentary at that level. But I will make a few brief comments on the trading conditions we're seeing in our specific markets.

The last 2 months has seen an encouraging pickup in extrusion and foundry activity in the U.S. generally. As you know, that's driven largely by the automotive and certain parts of the building and construction sectors. Same is true in Europe. There we're seeing a good pickup in billet, foundry, wire rod, other markets. Obviously in all these markets, the recovery is still uneven. You've obviously seen activity in China looking broadly encouraging.

All this said, customers particularly in the U.S., do remain somewhat cautious. Our run rate of value-added product orders in the third quarter thus far is showing only marginal pickup from Q2. However, there's been kind of an interesting trend over the last couple of months. We've seen very strong spot orders at the end of the month. To us that shows that our customers are exercising understandable caution, but also that actual conditions are stronger than they're expecting. It goes without saying obviously that the next couple of months will be subject to a lot of uncertainty.

The strengthening in these overall conditions is evidenced by an increasing array of product premiums. For example, the U.S. spot commodity billet premium now stands above $0.07 a pound. That was essentially 0 at the height of the crisis. The Midwest premium also improved from as low as $0.08 to its current level at just shy of $0.12. The freefall in the Midwest premium was evident well before the impact from the public health crisis. If you go back to May of last year, May of 2019, that of course is the date when Canada was exempted from the Section 232 tariff. And you measure the Midwest from that point to January, obviously before the impact of the pandemic, the Midwest was down 35% May '19 to January of this year. Of course the impact of the health crisis has brought it down further.

The collapse of the Midwest is very straightforward, the cause of it. It's due entirely to the surge of imports of primary aluminum from Canada. Canadian imports since the exemption in May of 2019 are up over 80% versus the same period before the exemption. At the same time, Canadian exports to other markets have fallen off a cliff. Exports to Europe have halved from an already low base. Exports to the rest of the world have fallen to nearly 0. This all runs directly counter to the commitment made by Canada at the time of the exemption, specifically that imports wouldn't surge in this manner.

As we've discussed at length, the tariff worked precisely as intended when it was effected. Three U.S. smelters restarted, with all the associated jobs and economic activity. U.S. production was up 60% versus 2018, that's '19 over '18. And there was no harm of any type to any downstream industry. No job losses, no metal shortages, no price inflation; in essence, none of the dire predictions that were heard at the time. In fact, 2019 was a banner year for the downstream industries. The only way to make this program effective again is the re-imposition of the tariff on Canadian imports, and we hope the administration will act to do this without delay.

Lastly, we continue working with the newly formed Goose Creek, South Carolina municipal utility on preparations for a new contractual arrangement for the Mt. Holly smelter. Regrettably here, the state-owned utility continues to obstruct and delay the process. And thus, we've got no choice but to engage in several legal courses of action. Time, of course, has marched on, and the next couple of months will be key in determining the future of this plant. As a reminder, Mt. Holly is the newest smelter in the U.S. It's got a well-earned terrific reputation as a quality billet supplier. It's got a truly great team of dedicated long-serving employees and it's got access to a plentiful supply of natural gas-fired power in the Southeastern part of this country.

The problem of course remains that the legacy power company continues to demand that Mt. Holly purchase 25% of our power requirements from them. The natural gas-fired power that we buy from the market, that 75%, is priced inside the median global power rate paid by smelters, better than the median. However, the price of the power from the legacy power company remains at over 2x the price of the power we buy from the market, and thus the weighted average of that 75% and that 25% is on the margin of the third and fourth quartiles of the global cost curve of smelters, making this plant unnecessarily uneconomic. We remain absolutely 100% committed to finding a way to make this excellent plant viable. Breaking this log jam with the legacy power company is the only thing that stands in the way and solving this would enable Mt. Holly to operate at full capacity both potlines, as it should.

And with that, I will give you over to Pete for some comments on the industry.

Thanks, Mike. If we can move on to Slide 4, please, I'll take you through the current state of the global aluminum market. The actual cash LME price averaged just under $1,500 per tonne in the second quarter, which was down approximately 12% or $200 per tonne from the first quarter, as COVID-19 weighed heavily on the global economy in the quarter. However, industry conditions continued to improve, and the LME prices averaged $1,640 per tonne for the month of July and the current price is just shy of $1,750 per tonne.

In the second quarter, regional premiums averaged approximately $0.09 per pound in the U.S., down 35% quarter-over-quarter, and approximately $100 per tonne in Europe, a decrease of 30% from the prior quarter. Current spot prices are around $0.12 in the U.S. Midwest and $120 per tonne in Europe. In the second quarter of 2020, global aluminum demand was down about 9% as compared to the second quarter of 2019. China showed strong signs of recovery in many end markets during the quarter, as evidenced in demand growth of 5% compared to the prior year quarter. In the world excluding China, we saw demand contraction of nearly 30% from the prior year quarter. However, there has been a sharp recovery in manufacturing activity in the world excluding China, particularly in the United States, as well as Europe.

Despite volatile price movements in the sector, global production was flat in the second quarter year-over-year. We saw a 2% production growth in China versus the same quarter last year, which was offset by a 1% decline from the rest of the world in the same period. With this backdrop, we've seen the LME price rally to its highest level in nearly 6 months, on the heels of a rising SHFE aluminum price, a weaker U.S. dollar and global manufacturing expansion led by the U.S., China and Europe.

Just a quick comment on raw material prices before I hand you over to Craig, the alumina price index has ticked back up a bit since the low of approximately $240 per tonne we saw during the second quarter. Currently, the alumina price index is about $270 per tonne, or less than 16% of the LME price today.

Okay, with that, I'll hand you the call to Craig.

Thanks, Pete. Let's turn to Slide 5, and I'll take you through the results for the second quarter. On a consolidated basis, global shipments were up 4% quarter-over-quarter, while realized prices were down 8%, primarily as a result of lower lagged LME and regional premiums.

Looking at operating results, adjusted EBITDA was $4.7 million this quarter and we had an adjusted net loss of $18.4 million or $0.19 a share. In Q2, the primary adjusting items were $6.4 million for the net realizable value of inventory and $2.7 million for unrealized impacts of forward contracts. Our liquidity remains strong with $197 million of funds available via a mix of cash on hand and credit facilities. As Mike mentioned earlier, we successfully refinanced our $250 million note on July 1, effectively extending its maturity out into mid-2025.

Concurrent with our refinancing, we also extended the maturity of our U.S. revolving credit facility into 2023. Let's turn to Slide 6, and I can walk you through our quarter-to-quarter bridge of adjusted EBITDA. Decreases in realized LME prices and regional premiums comprised the majority of EBITDA reduction versus Q1 levels. The Q2 realized LME of $1,630 per tonne was down $120 per tonne from Q1 levels, while realized Midwest premiums of $249 per tonne were down $55 per tonne over the same period. These actual realized values are slightly lower than those we discussed on our last call. Keep in mind that a proportion of our U.S. sales price on a 1-month lag, and when we provide these estimates each quarter there are volumes still unpriced. But we generally assume flat pricing for the remainder of the quarter.

As you know, the Midwest premium in particular was extremely weak in May, with an average level about 17% less than when we last spoke in April. Realized alumina prices were roughly flat quarter-over-quarter. The mix impact from the slowdown of billet and other value-added product demand was about $4 million versus Q1. As Mike mentioned, we've already begun to see the resurgence of this volume in the third quarter.

Domestic power prices continually dropped throughout the majority of the second quarter and generated a 7% or about $1.50 per megawatt hour savings versus Q1. As we discussed previously, approximately 30% of our Icelandic power pricing is now based on the Nord Pool index, which was down $11 per megawatt hour, or about 65% from Q1 levels. Finally, our sustained effort in controlling operating cost amid this uncertain environment was the primary driver of $5 million of increased earnings versus prior quarter.

Looking ahead to Q3 specifically, the lag LME of $1,530 per tonne is expected to be down $100 per tonne from Q2 realized prices. The Q3 realized U.S. Midwest premium is forecasted to be $240 per tonne, or down about $10 per tonne. And the European delivery premium is expected at $100 per tonne, or down about $40 per tonne versus the second quarter. Realized alumina is expected to be $280 per tonne, or down about $20 per tonne versus prior quarter. Taken together, the LME, alumina and delivery premium pricing declines are expected to negatively impact Q3 EBITDA by about $20 million to $25 million versus Q2 levels. Additionally, seasonal power price increases are expected to negatively impact Q3 EBITDA by about $10 million versus Q2. Please keep in mind that we buy in the day-ahead market and we still have 2 months of unpriced purchases assumed in this incremental impact. In sum, we expect these items in isolation will equate to an approximate EBITDA decrease of $30 million to $35 million from Q2 levels.

Let's turn to Slide 7, and we'll take a quick look at our cash flow over the last quarter. We started the quarter with $148 million in cash and ended June with $174 million. During the quarter, we had $4 million of CapEx spending, the largest individual component of which was related to the Hawesville restart and represented the final expenditures to support a newly restarted and rebuilt 4-line operation. Our normal semiannual note interest payment was a usage of $9 million, and working capital was a sizeable inflow for Q2, primarily driven by inventory reductions as a result of lower raw material carrying levels versus prior quarter.

Finally today, I'd like to provide some insight on how the recent trends in LME prices and Midwest premiums could impact our business in the future. As Mike mentioned earlier and as I'm sure you've all seen in your own research, LME prices have continued to increase rather steadily from the recent low point in mid-April of $1,425 per tonne to today's spot price of $1,740 per tonne. Midwest premiums have followed the same upward trajectory from their recent low point in early May of $176 per tonne, to today's spot price of $259 per tonne. Keep in mind that the Q3 commodity based sensitivities we shared earlier were derived using our normal lags across the portfolio, hence creating a timing disconnect from the recent market recovery to the anticipated P&L impacts. In other words, the recent market recovery in LME and Midwest premium will begin to impact our reported results in the fourth quarter.

Let's turn to Page 8, and I'll take you through a simple analysis we've put together to illustrate the potential impact of recent pricing trends on a typical quarter for the company. We begin with our Q2 adjusted EBITDA of $5 million. As we discussed earlier, this result includes a realized LME of $1,630 per tonne and Midwest premium of $249. Updating the LME and Midwest premiums to spot prices of $1,740 per tonne and $259 per tonne, respectively, resulted in adjusted EBITDA of about $24 million. Again, this is a representative example of the impact of spot prices on an actual period, and is not intended to be used as an outlook for any particular period. Movements in other prices from Q2 levels other than the LME and Midwest premium could also affect the sensitivity shown. In short, recent spot pricing, while still depressed versus historical levels, will benefit our P&L materially in the future due to the lagged nature of the LME and Midwest premium pricing.

This concludes our prepared remarks. Thank you for your time and attention. I'd like to turn the call back over to Brandi to begin the question-and-answer session. Brandi?

(Operator Instructions) Your first question comes from the line of Lucas Pipes with B. Riley FBR.

Just a few questions here from my side, and first I wanted to touch on Mt. Holly. It's like things are moving in the right direction there slowly but steadily. Can you remind us what the procedural steps are from here on out and how quickly we may have a (inaudible) there?

Yes, sure Lucas. It's Mike. I'll take that one. Thanks for the question. Without getting into the specifics of the various court cases, there is several here that are ongoing. Those will sort of wind their way through those processes over the coming couple months. The real -- the date that's in front of us, of course, is 31 December, which is the expiry of the current power contract. And so I guess I'd say over the next month or 2 or 3, we need to develop some confidence as to how the legal posture will look, as we look to a replacement for that power contract. So I guess it's a throw-away comment to say these processes take some time, unclear. It's very difficult to predict, and thus I won't try whether any of them will or all of them, obviously would be at their fruition over the next couple months. But we do have other alternatives on which we're working, and we will be making some of those decisions over the next couple of months. Ultimately, we think we've got a winning hand. It's just a timing issue.

Any additional color (inaudible)?

No, very difficult at this point. We want to continue running that plant. And ultimately, Lucas, we want to run both potlines, because that's the way it was designed and that's the way it should run. And the demand for the product, the billet, is there many, many, many times over. So that's the pot at the end of the rainbow is running the plant at 2 lines. It's running at one line today. And we're going to, as I said, have to make some decisions here coming up as to how we get from where we are now to the end of the rainbow. Specifically on your question, there's a bunch of court cases. There's cases in federal court. There's cases in state court. It's pretty complex. It really -- I don't -- without being presumptuous, it might sound that way, expounding on the details of any of those cases isn't going to do anybody any good.

I understand and I appreciate the color you were able to provide. My second question is I'll try to touch on a couple points, in kind of one swoop. The first is just kind of understanding this price recovery in Midwest premium specifically. You mentioned demand has declined 30%. North American production is up. You have the import surge from Canada. Kind of how do all of these pieces fit together? And then of course as it translates to your results, [we could be] looking for a much stronger Q4. Could you share kind of where we might be shaking out on an adjusted EBITDA basis for the fourth quarter?

Sure. You've got some -- I think in your question you've got some apples and oranges, and maybe a kumquat and a grape. So let me try to dissect it or parse through it. The 30% down, and I think that's in reference if I recall, from our first quarter call, was simply referencing not total primary aluminum demand decline, but decline in value-added products, the billet and other value-added products, billet slab, et cetera, et cetera. So that wasn't referencing the total [un-rod] aluminum or primary, however people use those synonyms. As you correctly referenced, the significant increase -- I think if I understood your question correctly, over the last couple of months and frankly over the last 15 months, has been from Canada, as you look at the import data. It's all up on Commerce's website. So obviously all those data are there. That's really been the difference. U.S. demand has been bumping along, but the significant proportion of imports that are coming down from Canada have significantly changed the market. I mean that 80%, for example, increase 12 months over 12 months, has driven the market share of Canada's imports in the U.S. from around 50% which was an all-time high, to just shy of 80% today. So it's just -- it's simply a source of where the demand is being fed from, as the Canadian imports have surged into the U.S. Does that address your question, or did I miss it a bit?

Very helpful for now, I'll let you get to the second part of the question. I may have a follow-up.

Yes, please jump back in the queue.

Yes, on the EBITDA, would you be able to touch on that, Q4?

EBITDA specifically, in what respect? What's your question?

The question is, so with the price recovery in LME, with the price recovery in the Midwest premium, obviously Q3 is going to be weaker. But then kind of as we think about Q4, given today's prices, what could EBITDA look like?

Go ahead, Craig.

Yes, that was still -- I'm sorry, Lucas, we're having a little trouble hearing you. That was still Lucas, right?

Yes, yes.

Okay, yes. Okay, Lucas. We're not going to talk specifically about Q4 today. We'll certainly save that for the next call. But I think a good way to do that, we did share a sensitivity based on Q2 actual today, which would give you kind of call it a mark-to-market for 2 of the largest movers. And then also in the appendix we've included our second half sensitivities on EBITDA, right? So if you were to take what your assumptions would be for LME and Midwest premium and add those onto an actual quarter, 2Q would be within a relevant range a good place to start. That would give you a sense.

Yes, I mean this is exactly why we don't -- one of the reasons, many reasons in my opinion, why commodity companies find it difficult to provide guidance and [probably shouldn’t] specific guidance. But as Craig has pointed out in detail over the last couple of years, given the lag, you can impart your own estimates on commodity prices 2 to 3 months on average, as we've said. We've given you the data as to how much of our sales are just 1 month lag. And given your own estimates for LME prices and delivery premiums, principally Midwest, but also the European Duty Paid Premium, you can work out your own math. I mean the math through which Craig took you was pretty simple. It answers the question simply taking Q2 and changing nothing but the LME realized and the Midwest realized to today's spot prices, what would Q2 have looked like pro forma? A bunch of finance words obviously, and that was what was on that last chart there.

Your next question comes from the line of John Tumazos.

Thank you for the explanation of the lag in your selling price. Is the lag on your alumina cost the same? Or is it even longer because the alumina has to be delivered and then worked through your work in process?

Great, great question. Thank you for the FIFO inference on the back of that. I'm going to let Craig answer that. The answer is it's kind of the same, but you get there in a different way. But go ahead, Craig.

Sure. No, that is a very good question. At the highest level, from a book perspective, right, from a book reported perspective, that's going to come through with about a 3-month lag. And that, depending on where inventory levels go, that could be on the -- I'll call it the south end of that, call it a 2-month. It could be as long as 4 months, but on average about 3 months. From a cash perspective, to your point, it's on a 1-month lag.

So that's the salient point John, is that unlike the sales, the vast majority of our alumina is priced we paid for it on a 1-month lag. That's the cash that's actually flowing out the door. But to your very good point, via the logistics, i.e. getting the stuff either oceangoing cargo and/or barges, and then depending on how much stuff we have in the siloes to Craig's point, that adds a weighted average with 1 to 2 months to that 1-month contractual pricing lag. Does that make sense?

Yes, thank you. If I could ask a second question, you made reference to the rebound in demand. The aluminum association orders for the first 6 months are down 15.6% from last year. Clearly housing is recovering really well. Wood prices are at records 45% higher than March. Are the aluminum customers liquidating inventory, or why is the rebound sort of not obvious from looking at orders?

Yes, there was some inventory destocking. It goes without saying, not just customers John, but throughout the supply chain, right? You had a lot of middle folks, whether they're distributors, or wholesalers or depending upon how the specific downstream sector is structured from a commercial standpoint. So clearly some of that went on. It's really hard to put your hand on how much of this is due to restocking to replace that destocking. But look, this 6 months, it's so hard generally in these markets, as you well know, and especially hard, given 6 months like we've just seen, the kind of sight that block of time and kind of infer anything from it. I think we need to all watch it. As you said, we need to watch auto builds. You saw that printed last week. We need to start watch obviously how these starts or applications for new housing permits, all the -- I think you guys know. You know this, all the forward -- so-call forward-looking indicators that are mortgage applications, et cetera, et cetera that are good indicators. It's kind of hard to tell. Let's have this -- I'm not trying to give you a non-answer. But this will be a more, I think, interesting discussion a couple months from now.

Within the aluminum orders, wire and cable was up 24%. Is there a particular reason why electrical transmission or wire and cable would be doing better? Is it more closely tied to housing?

It's tied to housing, but it's also tied to general infrastructure spending. Wire rod and wire and cable has been a reasonably, if I think of even the worst of times and I look at our specific customer base, those customers orders were literally spot on, no pun, dead on their contractual binds for the year, in fact, some of them exceeding. So at least in our small neck of the woods and then as you say referring to the broader industry data, that sector never took a pause. I mean utility spending does and did continue through whatever, including a pandemic.

94% of my town doesn't have electricity. There's got to be a demand for wire and cable out here.

I'm sorry about that. I hope you don't need, John, a roof or something like that.

No, I just went to the firehouse to charge my devices to hear your call, Mike. It's all good.

That makes us feel good, John. Thanks. Good luck, by the way.

Thank you.

Your next question comes from the line of David Gagliano with BMO Capital Markets.

I wanted to focus in a little bit, just on some of the funny math on Slide 8 sort of thing. So Q2 adjusted to spot LME and regional premiums is $24 million of EBITDA. And then the other adjustments I'm assuming are going to be power, which you mentioned is -- I'm assuming that's right. What you mentioned for the third quarter was about a $10 million likely drag. And then I guess the only other one would be alumina, which is if you use your sensitivities and use the 3Q bridge, math sort of thing would imply maybe a $4 million benefit, right? So that's kind of -- those are the other 2 main puts and takes? Is there anything else that we should be thinking about?

Go ahead, Craig.

Well, a couple of things, David, just to split those topics. So the first one, so we're crystal clear here, on Page 8 that is just marking our current quarter. So we took our printed results and just marked it to spot, to give an idea, right? So I just want to make sure for everybody else on the call too, we don't conflate that with looking at Q3 go forward. So if we go to Q3 go forward, so the 3 pieces we talked about, the first LME and regional premiums. I think you got those, right? So down about a hundred on LME, down 10 on Midwest premium, and down 40 on EDPP, right? Power, you were right, about down 10. Alumina, per the sensitivity if we look at it at about $20, it's going to be quite a bit less than [$4 million].

David, I think -- sorry. I think you were asking about other changes that (inaudible) to that Q2 pro forma, right? And so I mean power, power printed in Q2 pretty consistent with a typical Q2 spring shoulder season in power land, et cetera, et cetera. There's no reason if you look at the forwards and whatnot, there's no reason that that should be up or down. Obviously it's up, as Craig correctly said in Q3, just given the hot summer, a hot summer, a typical seasonal trend. And then allied, yes. I mean as we've said before, we've given you the sensitivity, but I think I'm pretty sure we talked about this last time. We have very little exposure for the balance of -- by design -- for the balance of 2020 for the API. And so we're buying almost on a percentage LME basis, which adjusts itself, if you will, for those sensitivities.

Got it. Thanks for clarifying that. And then just on the -- you mentioned sort of marginal project approvals -- or not marginal, sorry. That's not the right word, but modest project approval I believe was the word. Is there a potline restart, rebuild, restart in that mix?

No. But you're really hot on it. So the most -- the biggest one is, as we told you guys last time when we talked, one of the things that we did more for sustainability of operations reasons than economic reasons, because even at the dearth of commodity prices, a cell rebuilding at Grundartangi and Sebree was still measured in a simple cash-on-cash payback way, way, way inside a year. And so we restarted, for example, Sebree where we had stopped it. We started normal cell relining activity now. And so no, it's not a full potline. But it's just getting a couple cells that in normal times would have been relined and back in service within 6 or 7 or 8 days is our normal turnaround times for when they were cut out or failed, via the normal course. It's just getting those patched up and rebuilt and put back in.

Okay, and just the last question from me, the commentary, obviously there was a lot of focus on Midwest premiums and what's driving the runup. And obviously you're clearly focused on Canada, which makes logical sense from your perspective. And my question is, if you can provide us a little more insight into what you think is actually going on right now within the administration with regards to that issue. And secondly, fundamentally if there were to be tariffs reimplemented against Canada, fundamentally what do you think the Midwest premium should go to? Where do you think it should go to?

Yes, sure, David. On the first, the real answer is we don't know. We don't know any better than anyone else. But all we can do and what we do do, it goes without saying, is look at what they say. And so you know, the most recent tangible stuff that's come out is like -- probably goes back now almost a month perhaps, when Ambassador Lighthizer testified in front of Congress both House Ways and Means, and Senate Financing. He talked at reasonable length about this specific issue. So they're obviously very aware of it. That's in the U.S. Trade Rep's own words and are focused on it, and have identified surge and he said a lot of things like that. And so, but other than that, we don't know. We know that they agree it's a surge, based on investor-related comments, and that they're focused on it and not much else. In terms of where the Midwest goes, it's hard to tell. I mean you know where it was right before when the tariff was affect, i.e. before the exemption and before the surge started a month after the exemption and started to build and build and build and build and build. And then of course you saw where it was. It was in the sort of $0.16 to $0.18 plus category. And then it came down to $0.12-0.13. And then it came down further. We call that the pandemic effect, where beginning in February, it came down to $0.08. Where it goes? You know, some of it is LME-dependent of course, given the circular reference in it. But you can look back at where it was when it was affected and start to draw some inferences.

Your next question comes from the line of Paretosh Misra with Berenberg.

So first of all, just wanted to talk about the coke and pitch market, just curious what you're seeing there. I mean a lot of us think of coke and pitch as the derivatives of oil. So did those prices follow oil lower in Q2? And then could that be a sequential headwind as we look into Q3?

No, I mean you have 2 things going on there. Yes, crude prices are down, but on the other hand, the production of those products, specifically coke, which of course as you know, we use -- a smelter uses multiple times more than pitch, 4 to 5x more, is dependent on refinery runs. And so when demand for refined products goes down that's a problem obviously, there's less demand. The answer is coke picked up a little bit, but it's now -- where has it been, Craig? High 200s Gulf delivered?

We just closed the quarter at 230.

Yes, okay, 230. So it's still way below that. Pitch has been pretty sticky. It's been again, the reprice, 800 give or take, 790, thank you. So you could see some movement there. But they've both been, there's so many factors. You've got demand down for those products. You've got the price of the raw material down, as you correctly cited. But also the supply down due to the first factor, i.e. less demand for refined products. It's a bit of a mishmash of factors that affect those markets.

Thanks for the color. That's very useful. Then I just wanted to check. You provided cost guidance for the second half of this year in your last call. So I just want to make sure that you still expect to run fairly close to that guidance. I mean of course after adjusting for all the sensitivities, but otherwise are you on track?

We are, Paretosh. I would take that -- you know, you're right. So the sensitivity is we're going to see a little bit of a reduction in operating costs as some of those costs go down. And obviously use the LME sensitivity on the revenue side. But when we get to the SG&A, interest, our CapEx spend; very much in line with what we shared last time.

And I guess last one, and just expense, is there any change in that for -- given after the refinancing?

I'm sorry, interest expense, did you say, Paretosh?

Yes, that's right, yes.

Yes, so just I mean coupon, that's --

Yes, it will be nominally more.

It will be, you can do the math now, so 250. The coupon is all-in, including the pay-in-kind, the 2 points of pay-in-kind is 12%. So it will tick up $2 million a quarter, just quick math.

Your next question is a follow-up question from the line of Lucas Pipes with B. Riley FBR.

I wanted to get a little bit better sense for where you believe the Canadian imports are ultimately coming from. So Canadian kind of homemade production has picked up as well. Is that the lion's share of increased imports into the U.S. or is there more going (inaudible)?

No, it's -- yes, so it's twofold. One is yes. Of course there's been production restarted in Canada, just like there's been production curtailed in the U.S. at the same time as the production came on in Canada. But the other is just -- is a redirection of exports from Canada to other markets into the U.S. in order to capture the tariff. Again, this is exactly what was supposed to happen.

So it's a redirection of Canadian exports? It's not that -- have we seen additional imports into Canada?

Oh, no. It's all Canadian production. I'm sorry. Are you getting at -- are you asking -- perhaps I misunderstood you. Is there product coming from elsewhere through Canada? This is all Canadian production, absolutely. If I missed your question, I'm sorry.

No, no. It was -- you addressed it. I wanted to get a better sense of at the end of the day, where it comes from.

Oh, no. Canada. Canada. Canada. Yes.

(Operator Instructions) And you do have a follow-up question from the line of Paretosh Misra with Berenberg.

Just curious as we are entering August and typically order book slowdown or new order slowdown at this time of the year; are you starting to see that this year or not so much? And do you have any expectations as to how the summer might look like versus a typically seasonal slowdown?

Sure, I mean just to mark it back to whatever a typical environment for us, i.e. pre-COVID, as you know, the vast majority, 90%-plus-plus of our production is sold on a long-term contract basis. Long term meaning 1 year-plus, and so we just don't have a lot of sales in the spot market. And so the takeaway there is our order patterns are pretty, in a normal environment, are pretty stable month to month, unless a customer has any specific seasonality that's built into our production forecasts anyway. Here, as I said, we're actually seeing a little bit of an encouraging uptick where customer when we were in the depth of April-May, even come up through June, sort of indicated where they thought they would be versus their contractual quantities for June, July, August thus far. And we're seeing sort of better than that. And so that's why I said, I don't blame those manufacturers. They're exercising caution. But business seems to be a bit better than they had planned, because we are seeing more spot orders than we would normally see, specifically of course for value-added products. We're talking about billet.

And there are no further questions at this time.

We thank you all, as usual, for the time, and look forward to talking with you in a couple months. Take care.

This concludes today's conference call. You may now disconnect.