Arch Resources Inc (ARCH) 2018 Q1 法說會逐字稿

Good day everyone, and welcome to this Arch Coal Incorporated First Quarter 2018 Earnings Release Conference Call. Today's call is being recorded. At this time, I would like to turn the call over to Ms. Logan Bonacorsi, Director of External Affairs. Please go ahead, ma'am.

Good morning from St. Louis. Thank you for joining us today. Before we begin, let me remind you that certain statements made during this call, including statements relating to our expected future business and financial performance, may be considered forward-looking statements according to the Private Securities Litigation Reform Act. Forward-looking statements, by their nature, address matters that, to different degrees, uncertain. These uncertainties, which are described in more detail in the Annual and Quarterly Reports that we file with the SEC, may cause our actual future results to be materially different than those expressed in our forward-looking statements. We do not undertake to update our forward-looking statements whether as a result of new information, future events, or otherwise, except as may be required by law.

I'd also like to remind you that you can find a reconciliation of the non-GAAP financial measures that we plan to discuss this morning at the end of our press release, a copy of which we have posted in the Investor section of our Website at archcoal.com.

On the call this morning we have John Eaves, Arch's CEO; Paul Lang, Arch's President and COO, John Drexler, our Senior Vice President and CFO, and Deck Sloan, our Senior Vice President of Strategy and Public Policy. We will begin with some brief formal remarks, and, thereafter, we will be happy to take your questions. John?

Thanks, Logan. Good morning, everyone. I'm pleased to report that Arch made excellent progress on a number of key objectives during the quarter just ended. First, we continued to expand our reach in the seaborne coking coal markets, cultivating and expanding our international customer base and increasing our export volumes. Second, we achieved our strongest-ever cash margin in our metallurgical segment at $48 per ton despite cost pressures in the portfolio - put another way, the margin that represents over 40% of the selling price.

And third, we bought back another 400,000 shares of stock at a price of $39 million, bringing our total buyback since May 2017 when the program was first initiated to $341 million. In fact, we've now bought back roughly 18% of our shares outstanding at a time of the program's initiation, in the span of less than 12 months. The size and speed of those buybacks is a clear testament to Arch's substantial cash generating capabilities, as well as our strong commitment to maximizing value for our shareholders.

As everyone on this call knows, we operate in a highly complex business environment, which can lead to significant degree of variability in our quarterly results. The first quarter 2018 was no exception. During the period, we encountered operational challenges at our Mountain Laurel mine that pushed up our metallurgical segment cost, as well as a series of logistical challenges that acted to dampen otherwise strong financial performance. Paul will discuss these issues further in his remarks.

However, I'd like to note that our increased emphasis on seaborne coking coal markets is likely to result in more timing-related issues associated with ocean vessel movements. Despite these issues, we had a solid performance overall as we again generated more than $100 million of EBITDA for the quarter.

Looking ahead, we anticipate robust free cash flow generation as the year progresses, with modest maintenance capital requirements, exceptionally low interest expense, and effective tax rate of 0%. That free cash flow outlook should facilitate a continuation of our proven and effective capital return program, which we believe is a powerful driver of shareholder value.

Turning now to the marketplace, we continue to see coking coal fundamentals in a strong and supportive light. Steel demand is up more than 4% year-to-date. Steel prices remain strong in our core Atlantic Basin marketplace, and global economic growth remains robust. The upshot is that coking coal prices remain buoyant. At present, [Platts] is assessing the East Coast price of High-Vol A coal at $185 per metric ton. While that's down somewhat from recent highs, it's still translating into a very attractive margin for Arch's primary coking coal products. Moreover, High-Vol A continued to trade at a significant premium to Low-Vol throughout the quarter, reflecting its relative scarcity in the marketplace. As of today, that premium was assessed at around $15 per ton.

On the supply side, Australia, the dominant supplier of seaborne coking coal, appears to be headed for a stronger shipping year in 2018. In particular, second quarter 2018 volumes are likely to be up strongly versus last year's cyclone-impacted year. Still, questions remain about the ability of rail and port system to move the increased volumes in 2018, and several large producers are guiding to flat or lower volumes for the year due to a range of geological and operational issues.

On the thermal side, the story's not changed much. Seaborne markets remained strong, with API-2 prices in Northern Europe above $80 per ton, and Newcastle, Australia price is above $90 per ton. That translates into attractive net-back prices for our West Elkin Coal-Mac volumes. This also means that incremental U.S. volumes will continue to be pulled in the seaborne market, which should translate into a more advantageous supply-demand balance domestically.

Meanwhile, generator stockpile correction is entering its 10th quarter, and there's still work to do. On the positive side, stockpiles were reduced by an estimated 10 million tons during the first quarter, and currently stand at around 70 days of supply. That's still 10 days or so too high, but it is progress. Looking ahead, we believe that the outlook for continued value creation for the balance of the year is bright. We're sharply focused on capitalizing on strong coking coal market, controlling cost across all three operating segments, and taking our thermal products to the market in a way that optimizes the value of our reserves. Moreover, we continue to believe that we have an excellent plan in place for driving shareholder value via our proven and highly effective capital return program.

In closing, I want to thank our employees for their exceptional work throughout the quarter. Simply put, our success rests on their talent, commitment, and passion for excellence.

With that, I'll now turn the call over to Paul Lang for more detail on our operating and marketing results. Paul?

Thanks, John, and good morning, everyone. Today, I'd like to provide additional color to John's broader market comments by digging deeper into Arch's operational results, as well as our sales and marketing tactics during the quarter.

Starting with the metallurgical segment, we're extremely pleased with the results of our decision to move more coking coal volume into the seaborne markets. In fact, our coking coal realizations were up nearly 30% over the fourth quarter of last year. Quarter-over-quarter, our segment cash margin increased more than 50% to about $48 per ton, leading to our strongest-ever cash margin on record for this segment. These strong pricing levels were driven by increased seaborne coking coal shipments, higher pricing on index-linked and negotiated business, and better fixed pricing on annual North American commitments as compared to 2017. It's important to note that our first quarter coking coal realization included approximately 200,000 tons of North American carryover volume. This largely completes our fixed price legacy commitments from last year.

Beyond the carryover volume, we shipped 1.3 million tons during the quarter, with nearly 70% of that volume priced on an index basis at an average netback of approximately $155 a ton. The quality mix on our quarterly deliveries was largely representative of our production profile. In general, coking coal shipments during the quarter were less than ratable due to the previously announced weather-related logistical issues in January and seasonal restrictions for our Great Lakes customers. We're continuing to see some vessel slippage in part due to competitor volumes not reaching the ports on a timely basis, contributing to delays.

Since our last update, we have committed approximately 800,000 tons of additional coking coal for 2018 delivery. Of that total, barely 530,000 tons are linked to an index or subject to mutual negotiation, and the remaining 270,000 tons bears a fixed price of $125 per ton. A vast majority of these fixed price volumes are associated with spot seaborne sales that are likely to shift in the second quarter.

Looking ahead, we now expect to sell between 6.3 million and 6.7 million tons of coking coal in 2018, representing a 3% reduction from our previous midpoint. This revised target reflects the ongoing variability in seaborne shipment timing. Our cautious outlook for Mountain Laurel, as well as a conservative operating stance on our 2018 production at the Leer mine. It's important to note that we don't expect our remaining coking shipments to be perfectly ratable for the remainder of the year. In fact, we anticipate second quarter coking coal shipments to be in line with the first quarter volumes due to the scheduled long-long moves at Leer and Mountain Laurel, as well as a shipment schedule that is currently concentrated towards the end of June. As you know, and based on past experience, this could result in volumes sliding into the third quarter.

For the year, we now have 5.1 million tons of coking coal committed, of which 2.3 million tons are subject to a market-based pricing mechanism and 2.8 million tons are an average fixed price of nearly $118 per ton. At the midpoint of our revised guidance, we have roughly 1.4 million tons still uncommitted, with a majority of that open position being our High-Vol A coal destined for the seaborne market. We expect the quality mix of our coking coal sales will be 55% High-Vol A, 30% High-Vol B, and 15% Low-Vol in 2018.

As John mentioned, on the cost side, the metallurgical segment turned in a mixed operating quarter. While our Leer mine delivered a better-than-expected performance, Mountain Laurel again encountered tough conditions associated with the sandstone intrusion in the current longwall panel. As indicated, the operation is preparing to move to the next panel, likely late in the quarter, where we believe mining conditions will be more favorable. However, considering this recent development, we are tempering our expectations for the year at the operation.

Outside of the operating issues noted, we also incurred increased sales-sensitive costs related to higher coking coal prices, as well as several one-time expenses that hit during the first quarter. Together, these items accounted for more than 50% of the quarter-over-quarter cash cost increase. Going forward, we're now targeting an average cost structure in the range of $60 to $65 per ton for full year 2018. While this is a level higher than originally anticipated, we continue to view our portfolio as one of the most cost-competitive in the U.S. metallurgical coal sector.

Turning to the thermal franchise, in the Powder River Basin, customer shipments were flat when compared to the fourth quarter of 2017, supported by favorable winter weather that persisted for most of the quarter. However, ongoing uncertainty and future burn forecasts, still-high coal inventory levels, and lower pricing for competing fuel sources, led to muted buying activity during the period. In fact, we elected to place only one million tons of additional 2018 coal sales at an average price of approximately $12.60 per ton, representing a premium to the current physical spot pricing levels.

Given the current market fundamentals, we've made the strategic decision to reduce the production level at our Black Thunder mine. We now expect to produce between 62 and 68 million tons from the operation in 2018, a roughly 10-million-ton reduction from the midpoint of our most recent guidance. This approach will allow us to keep the tons in the ground for future periods rather than sell them for an inadequate return. We believe this measured marketing strategy for our Powder River Basin franchise is in the best long-term interest of the company and the shareholders.

Despite targeting a lower volume level, we remain keenly focused on holding the line on costs as we progress through the balance of the year. As a result, we are maintaining our previous cash cost guidance for the segment at $10.45 to $10.95 per ton for 2018. It's important to keep in mind that, due to the normal seasonality of Powder River Basin shipments, we'd expect the second quarter to be our lowest shipping and highest-cost period of the year.

With the new sales commitments signed in the quarter and our decision to lower the production level at the mine, we have reduced our uncommitted 2018 Powder River Basin volumes to approximately five to seven million tons, all of which is at Black Thunder.

Moving on to the others thermal segment, shipments were slightly below the fourth quarter results on planned lower volumes from our West Elk mine. Both West Elk and Coal-Mac continue to participate meaningfully in the export thermal markets as attractive international pricing and strong market conditions pulled U.S. coals into Northern Europe, Asia, and South America. For the year, we expect exports from this segment of more than 4.2 million tons, which is over a 50% increase from our 2017 levels. For 2018 deliveries, we committed 550,000 tons at an average price of around $34.50 per ton. We now have 8.3 million tons sold for the segment, all of which is priced.

As expected, the cash costs at our other thermal segment trended up in the first quarter, reflecting a smaller contribution from the lower-cost West Elk mine and a higher relative contribution from the higher-cost Viper mine. Given the anticipation of a slight increase in volumes from the segment, and ongoing cost control initiatives, we have slightly decreased our 2018 cash cost guidance for the segment to a range of $27 to $31 per ton.

Overall, we now expect our 2018 thermal shipments for the company as a whole to be between 80 to 84 million tons. Based upon the midpoint of this guidance, we are approximately 90% committed for the year.

Once again, our first quarter results demonstrate that the scale and range of our asset base is a competitive advantage. As we progress through the remainder of the year, we'll remain focused on reducing costs, executing on our seaborne-centric marketing approach, which continues to pay dividends, and increasing the profitability in each of our operating segments.

I'll now turn the call over to John Drexler, Arch's CFO, for an update on our financial position. John?

Thanks, Paul, and good morning. During the first quarter, we continued to execute on our plan to generate healthy cash flows, maintain a strong and flexible balance sheet, and execute aggressively on our capital return program. During the quarter, cash generation continued to be strong, with [66 man hours] of operating cash flows despite the impact of negative working capital adjustments of $70 million. The negative working capital adjustments stem in part from the logistical challenges and missed seaborne vessel shipments that occurred during the quarter. We would expect our working capital adjustments to moderate over the course of the year. Despite the operational and supply chain challenges this quarter, all of our operating segments continued to make solid contributions, demonstrating the capability of our low-cost asset base.

As John indicated in his remarks, we continue to return capital to you, our shareholders. Since the program's inception approximately a year ago, we have returned $341 million by repurchasing 4.3 million shares, or approximately 18% of the shares outstanding at an average price of $77.71 per share.

During the first quarter, we bought back an additional 407,000 shares for $39 million. As of March 31, we had approximately $159 million of capacity remaining under the $500 million repurchase authorization. In addition, the Board of Directors has approved the next quarterly dividend payment of $0.40 per common share. That dividend will be paid on June 15th to stockholders of record at the close of business on May 31. When combined, we have returned $373 million in just 12 short months.

Looking ahead, we expect to continue to execute on our plan of returning excess cash to our shareholders, and we will constantly evaluate which uses of cash provides the best risk-adjusted return over the longer-term. With this deliberate process, proven strategy, and expectation of strong free cash flows for the remainder of the year, we expect to further our progress on our capital return program over the course of 2018. The capital return program remains an important focus for the company and our Board.

We ended the quarter with $433 million of cash and short-term investments, well within our expectation to maintain a cash position of between $400 million and $500 million. We also ended the quarter with $329 million of debt, leaving us in a net cash position of $104 million. We continue to remain intensely focused on maintaining our industry-leading balance sheet and ample liquidity. We recognize that we operate in an industry that will experience cycles, and, as we have learned, a strong balance sheet provides crucial support through each phase of the market cycle.

Also in early April, we took another step in lowering our cost of capital further. On the back of ratings upgrades from both S&P and Moody's, we repriced our $300 million term loan, reducing the interest rate from L plus 325 to L plus 275. The rate reduction will lower our cash interest expense by $1.5 million on an annual basis.

Since our reorganization, we have refinanced and repriced our debt three times, taking an initial rate of L plus 900 down 625 basis points to L plus 275, and reducing our annual interest expense by $22 million. In addition, we have entered into interest rate swaps for a portion of the term loan, fixing the floating rate for the next several years. Details are disclosed in our security filings, but for the remainder of the year, we have swaps in place for $250 million of principal at an interest rate well below current market rates.

Our 2018 guidance is reflected in the press release, and Paul has provided thoughts on our sales and operating cost outlook for 2018. A few additional items to note - we now expect our SG&A expenses to be between $87 million and $92 million. This includes $15 million of noncash equity compensation expense. This increase of $3 million at the midpoint from last quarter primarily stems from one-time compensation-related expenditures and legal expenses during the first quarter. Our interest expense guidance of $16 million to $18 million reflects the impact of the term loan repricing, and we continue to expect our tax provision to be zero for the foreseeable future.

Wrapping up, we continue to believe there is great strength in our coking coal and thermal coal franchises. In fact, we are confident this expertly-paired operating portfolio, pristine balance sheet, and clear and systematic approach to capital allocation will continue to generate significant value for our shareholders.

With that, we are ready to take questions. Operator, I'll turn the call back over to you.

[Operator Instructions.] Mark Levin, Seaport Global.

My first question is on met-coal cash cost, obviously elevated, and you referenced what's been going on at Mountain Laurel. When you think about 2019, obviously 2018, you've run into a sandstone intrusion. You've had lots of logistical-related issues, well-documented. But when you think about the cost structure maybe looking out a year, is there anything that has changed in terms of the way you look at what the cost are for the met enterprise as a whole beyond the issues that you mentioned in '18? Put another way, can we go back to where we were before?

2019 is a little bit out, and I think it's a little premature to speculate too much on it. But I think, as you listen to what we've said in the past, we are expecting better conditions, getting back into thicker coal and longer panels at Leer. That'll happen. And the geologic issues at Mountain Laurel, that's obviously been a little more vexing for us. With that, though, I think I always get a little humbled by these Central App longwalls, particularly Mountain Laurel, but I think we have quite a bit of hope of getting back to a more normalized cost structure.

Is Mountain Laurel core? I know it's an asset that's been around for a while, and it feels like there've been issues there from time to time over the years. Is that an asset that -- I guess the political answer is, if somebody offers the right price, everything is for sale. But just in general, it just feels like this asset tends to create some issues, or has created some issues, over time.

[Six] months ago, I think I made the comment Mountain Laurel's had a great run, but obviously it's no longer in the elite status it was 10 years ago. Obviously, we mined our best coal first. But having said that, we're still expecting some significant volume contribution from the mine, particularly in 2018. And look, it's got a middle-of-the-road to higher as far as the competitor cost structure. And I think, unfortunately, we ran into a similar condition this quarter as we did in the summer of last year. I think we are expecting better conditions in the next panel, and we've not seen anything in development. But if the geology and mining costs stabilize, which is what we're shooting for, I think Mountain Laurel could remain a significant part of our portfolio. I think it's in the case where we're at now with that mine, is we're going to have good quarters, and we're going to have bad quarters. But I think as you look at it, the mine as a whole has a good international standing as far as coking coal. It's been used by steel makers all over the world. And I think it's clearly worth the struggle right now. It's a well-known brand, and we have a great management team there, so I'm hopeful, but it's very wide-eyed as we go into this year about that operation.

Let me just tag onto what Paul said. Certainly, it was a tough quarter for Mountain Laurel. You've heard me say many times, we're always looking at our portfolio and trying to make revisions to make sure that we've got assets that are creating positive EBITDA on a long-term strategic basis. If you'll remember last year in the third quarter 2017, we finally came to the conclusion with Lone Mountain that it just wasn't contributing enough. It wasn't strategic, so we decided to sell it. So this management team's always looking at our portfolio. We're willing to make the tough decisions. I don't think we're there on Mount Laurel at this point, but, longer-term, if we have an asset that's not making a contribution, we've got two options - we can close it, or we can sell it. And I think this team has been willing to do that. But as Paul said, we're going to move the longwall later this quarter. The next panel looks better. We'll manage it and see where we are. But if you look at our overall cost structure, we're still on the extreme low end of the cost structure in Central App and feel like we'll stay there.

Just the last question, I guess this might be more to John Drexler, when we're thinking about Q2. And I know you don't give quarterly guidance, but I was just trying to tick off some of the positives and the negatives in Q2 at least relative to Q1. I think you referenced flat met shipments, maybe the potential for some vessel shipping into Q3, weaker PRB shipments due to seasonality. Obviously met prices have come off Q2 versus Q1. So is it reasonable to expect, obviously, that Q2 would be down potentially significantly from Q1 before maybe recovering later in the year? Is that the right way to think about it, just the puts and takes?

Yes. I think, Mark, as you look at how we've laid things out, right, you've highlighted what we've identified in the release, that the second quarter is going to be in the PRB the lightest quarter from a shipment perspective. And I think we've been very clear also with the volume expectations in the met portfolio, as well. With all that being said, I think we continue to believe that we're going to be generating substantial cash over the remainder of the year, and the portfolio is well-positioned to operate here as we work through the remainder of 2018.

Let me just tag on a little bit here. We do have two longwall moves during the second quarter. We're building inventory at the mines, obviously. But when you look at our vessel schedule, we've got quite a few vessels that are scheduled in the back half of June. And what we've found with that shipping schedule, some of those could slip into the third quarter. So I think we're trying to be measured as we look at the second quarter, but certainly volumes should be stronger as we think about third and fourth quarter.

Has the logistical situation improved materially, not a little? How would you characterize what you've seen as we've exited winter? Is it getting any better, or has it gotten a lot better? How would you characterize it?

I'll take the first shot, then see if John wants to add. But from what we've seen, CSX has really come around. They've pretty well got themselves back in line. The NS, on the other hand, is still struggling pretty hard. Now, most of our impact has been on the thermal side of our exports because of that. And the third items I think I'd point out is where we see a lot of delays is when we're loading partial vessels, and by that I mean if we're -- we sold maybe three of the holds in a six-hold vessel, and the other party has not filled their obligation, or doesn't have the coal, our vessel has tended to sit. So yes, I think things are working out, but there's clearly been some growing pains the last six months, especially on the export side.

Lucas Pipes, B. Riley FBR.

I wanted to follow up on the cost side as well, and I thought I would kind of start with Q1 versus full-year guidance. So I think in Q1 you were at $68. Full-year guidance is $60 to $65. Obviously, that's cap met. And then in the second quarter, you alluded to two longwall moves, so I would think, all else equal, costs should maybe rise in the second quarter. So should we think second half of this year is substantially lower from where you were in the first quarter? And I would appreciate your thoughts around that. Thank you.

As you look at the first quarter, the costs in the $68 range, and if you look quarter-over-quarter, there were really three things that -- or two things that really brought that in. First, obviously, was a good one. We had a impact of sales sensitive, which was about $1.50 a ton. The second was we had a series of what I'd call unusual adjustments in Q1. There were things like a big subsidence claim we had at Mountain Laurel. We had a worker's comp adjustment. Those things are one-time, but they accounted for about half the quarter-over-quarter increase. So if you look at Q1 normalized, it was about $64. I think that's probably in the range of what you're going to see, slightly higher because of the longwall moves, then normalizing back to our guidance range in the back half of the year.

Maybe to hone in a little bit on Mount Laurel, when I look at [MSHD] data, it showed at about 450,000 tons of production in the first quarter. And that was a little bit lower from the fourth quarter, but also much better compared to the second and third quarter of last year. So what's the right run rate from here on out on Mount Laurel? And if you could maybe give a little bit more perspective as to the specific issues that you had, I would appreciate your perspective. Thank you.

Yes, Lucas, one of the issues with the MSHD data, as you know, is it is on a produced ton and not a sold-ton basis. So what you saw in Q1, particularly for us, is the met segment had a pretty good run rate as a whole, but we built about 240,000 tons of inventory in the quarter. And I think John alluded to that in our working capital adjustment. Now about 140 of that or 170 of that, I can't remember exactly which, was metallurgical coal, and the balance was [mids], so you saw some of that at Mountain Laurel. And as you look, going forward, at Mountain Laurel, I think we're still in that 1.8 million to two million ton range if we can sort out this issue that we've been having.

John Bridges, JPMorgan.

I suppose the good news is the High-Vol A price is still holding this premium over Low-Vol. Just wondered if you could give us a bit of color as to why you think that's happening to begin with.

Certainly, we think there's scarcity value in that product. If you look at the production on High-Vol A particularly has been taken out over the last couple years, whether it's U.S. or even Australia. There's just not a lot of people producing High-Vol A and low-cost High-Vol A. So we've seen that premium kind of move back and forth, but if you look over the first quarter, we've seen a $15, $20 premium typically between the High-Vol A. And going forward, we just don't see a lot of additional High-Vol A coming into the market. If you look back over the last year or two in all the new buy and coming into the market, more of that was Low-Vol and Mid-Vol, and not High-Vol A. So we think we're positioned well with not only the Leer production today, but if you think about the 130 million ton Taggart Valley reserve we have, we've got tremendous organic growth opportunities up there. Not in any hurry to bring that on, but we think it's pretty unique that we actually can replicate, with two additional longwalls up in that valley with low-cost, High-Vol A coal. Paul, you got anything to add to that?

Yes, I'd make the same comment. The High-Vol A is still very popular in Europe. It's very good for their blends. But, I think, surprisingly, there's still a very good market [over there] in Asia, particularly in Japan and Korea.

John, I'd add that it really, in 11 of the last 20 quarters, actually High-Vol A has traded at a premium. So we're starting to see this pretty consistently. And as Paul said, we really think there's great value and use for the High-Vol A and real asset type for that in Europe in particular, but increasingly globally.

I guess I'm going to have to change my models, go from a 10% discount to a 10% premium. So the logistics problems, could that be contributing to the strength of the High-Vol price? Because I'm just thinking that you've seen more volumes going out of Chesapeake Bay than is going out at the moment without these apparent choke points.

Look, John, I think that's a factor into this, is that the ability to get that coal offshore has been strained a little bit.

But John, as John pointed earlier, it's scarcity value across a wide range. So logistics have been strained in a number of places, but again, just really that lack of High-Vol A availability, and we're just not seeing more of it come online. And in fact, some of the High-Vol A mines that are out there have struggled with costs and with production issues. So we feel pretty positively about the outlook.

And John, the logistical challenges I think will hit all the products, not just High-Vol A, but it would be catching your Low-Vol and your High-Vol B, as well as PCI.

Just finally on Mountain Laurel, as you said, you take the best call first. So do you say you've got to be moving into another seam? What should we expect from the character of the coal, going forwards? And is part of the inventory issue a difference in yield of High-Vol B coal that you're producing from that asset?

John, the panel we're going to is adjacent. That's why I remain a little bit cautious on our outlook for the rest of the year. And we've actually, frankly, dropped our guidance a little bit on that issue. As far as quality in that and yield, it's about the same. Frankly, the biggest issue with these sandstone intrusions is really what it does to get through them. It just tears the equipment up. The shear in that is not made to cut solid rock. And we get through these intrusions, and we fight the maintenance issues usually for six or eight weeks after that.

Michael Dudas, Vertical Research.

John, you should be commended on your pullback on your PRB volumes. And so, on top of that, thinking about -- and you, of course, should be commended for being so open going into 2018 on the met-coal side. How were you thinking about this broad marketing strategy as you looked into the second half of the year into '19 as the U.S. negotiations come up on your positioning? Is it going to be different relative to what you had in the past, given where the market prices are relative to expectations in the past from the market or you guys? And what do you need to see before you want to get PRB back to a better run rate volume? Obviously an inventory day level would be nice, but maybe you can share a little bit of what you're thinking through there.

I'll start that question. As far as the met going into the future -- and I think we've been very honest in the past -- I think we believe North America's an important customer base for us for the met segment, but at the same time, we're not willing to accept a huge discount to stay in that arena. And going forward, I think we're just going to play it as it comes, and if there is an expectation of a discount in the U.S. that's more than the value, and that value is easier logistics, we'll stick to the international market.

Michael, I think it's a credit to Paul and his team. We did a pivot. We went from '17 to '18, and we went from 50-50 to about 80% of our product being the international market, 20% domestic. If you look at our portfolio, the products that we offer into the market, our logistical options, I think it allows us to be pretty nimble. So to Paul's point, we're going to evaluate the market opportunities later this year and really go with where the best economics for this company are. And fortunately, we have the ability to react like that fairly quickly.

And John, on the Powder River Basin, just to come to that side of it, as we look at where stockpiles are today, we think they're sitting at around 70 days. That's still more than enough, obviously. Now, we have seen some RFP activity that's been somewhat encouraging, but the fact is that there's still too much coal sitting in inventory. There's not enough demand pull. And so, as indicated, we were comfortable pulling back to a lower level of production, and we think that's the right place to be. Natural gas prices are sitting at 275, [that sits] at 275 right now. Given that, and given that outlook, we're comfortable with these lower levels of production. But obviously, if we see meaningful movement in demand and pricing, we can reevaluate.

And maybe a follow-up on the domestic versus international mix. Any thoughts on 232 aftermath, and what indications [you'd] be hearing from some of your current or anticipated U.S. customers for on the coker or the met-coal side?

Michael, it's Deck again. And certainly, we continue to feel fairly positively about Section 232. We think this administration is trying to find the right balance here to advantage the U.S. industry without precipitating too much contentiousness in the international arena, because obviously the international market is important to us. Clearly, higher steel production in the U.S. directly benefits U.S. metallurgical coal producers. And even if it's not us selling into that U.S. market as much, it is our competitor. So that's advantageous, and if we see increased production of steel in the U.S., that's a direct benefit. We are seeing indications, obviously, of some additional steel output and expectation for increased output as the year goes on, as you should see. The restart in Granite City being contemplated, maybe even the Ashland Works. And so, as we look at that, we do see that as relatively positive. Meanwhile, it does appear the administration is trying to work with some of our larger trading partners to make sure that they are not excessively disadvantaged, and those trading partners are important to us, obviously, as we are selling into the Atlantic Basin, selling most of our metallurgical coal into the Atlantic Basin. So we feel relatively positively at this point, but we'll continue to watch it.

It's all about making deals in Washington, Deck, isn't it?

That does conclude our question and answer session for today. I would now like to turn the call back to Mr. John Eaves for any additional or closing remarks.

I want to thank you for your interest in Arch Coal today. This management team is laser-focused on managing our costs, managing our balance sheet, being nimble on the marketing side, and executing on our capital return program.

Again, we thank you for your interest. We look forward to updating you on second quarter results some time in July. Thank you.

Ladies and gentlemen, this concludes today's call, and we thank you for your participation. You may now disconnect.