Ramaco Resources Inc (METC) 2018 Q4 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Ramaco Resources Fourth Quarter 2018 Earnings Conference Call. (Operator Instructions) As a reminder, this conference call is being recorded.

I would now like to introduce your host for today's conference, Chief Accounting Officer, Michael Windisch. Mr. Windisch, you may begin.

Thank you, Josh. On behalf of Ramaco Resources, I would like to welcome all of you to our fourth quarter earnings conference call. With me this morning is Randy Atkins, our Executive Chairman and CFO; Mike Bauersachs, our President and CEO; and Chris Blanchard, our Chief Operating Officer.

Before we begin, I would like to share our normal cautionary statement regarding forward-looking statements. Certain statements discussed on today's call constitute forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995. These forward-looking statements represent Ramaco's expectations or beliefs concerning future events, and it is possible that the results discussed will not be achieved. These forward-looking statements are subject to risks, uncertainties and other factors, many of which are outside of Ramaco's control, which could cause actual results to differ materially from the results discussed in the forward-looking statements. Any forward-looking statements speaks only as of the date on which it is made, and except as required by law, Ramaco does not undertake any obligation to update or revise any forward-looking statement whether as a result of new information, future events or otherwise.

New factors emerge from time to time, and it is not possible for Ramaco to predict all such factors. When considering these forward-looking statements, you should keep in mind the risk factors and other cautionary statements found in the company's filings with the Securities and Exchange Commission, including our annual report on Form 10-K. The risk factors and other factors noted in the company's SEC filings could cause its actual results to differ materially from those contained in any forward-looking statement.

With that said, I'd like to introduce Randy Atkins, our Chairman and CFO.

Thanks, Mike. As always, I want to thank everyone for joining us today to discuss our fourth quarter and year-end results. I also want to share some remarks on our 2019 guidance. As an overview from where we sit today, we are very positive about Ramaco's long-term prospects for growth, our capability to generate significant free cash flow, and ultimately, our ability to return cash to shareholders by dividend. We also remain in an aggressive growth mode to basically double the size of our company over the next 3 to 4 years, but let's start with 2018.

As we look back on Q4, the defining event was the silo failure at our Elk Creek prep plant in early November. As you know, we were able to get back to substantial throughput with the belt workaround by early December. We are ramping back to full capacity by Q2 of this year. Needless to say, not being able to wash and sell coal for 3 weeks negatively impacted Q4. Revenue dropped about 30% quarter-over-quarter to $44 million, and we went from an $11 million adjusted EBITDA in Q3 to only $7 million in Q4. For the full year, however, we had strong positive results both financially and operationally, certainly, in comparison to 2017.

Revenue rose about 270% to $228 million on about 1.75 million tons produced. We had almost a $52 million positive swing in EBITDA to $42 million and net income similarly rose by over $40 million up to $25 million in 2018. We spent about $48 million on CapEx in 2018. And on a normalized basis, we expect that number to drop in '19 by about 25% to somewhere in the $35 million to $40 million range. That figure is inclusive of the CapEx needed to get our Elk Creek prep plant back to full capacity. There are always going to be unexpected issues in the coal business, but had we not had the silo failure at Elk Creek and the sandstone washout at Berwind, we would have had a much stronger normalized 2018 results.

A bit later, Mike Windisch will review some accounting results for the quarter and year-end as well. As we look ahead to 2019, we have started strongly and we're predicting a materially better year than 2018.

Production at Elk Creek, we estimate at roughly 1.8 million tons. We have one more year of lower-volume development mining at Berwind at about 250,000 tons this year until we hit the lower cost and thicker Pocahontas #4 low-vol seam by mid-year 2020. Combined, our 2 mines should put us over 2 million tons of production overall for '19, which is about a 14% bump from last year. We also hope to sell over 2.2 million tons as we work down some large stockpiles from Q4 at Elk Creek.

We currently have about 175,000 tons of clean coal equivalent inventory at Elk, and we'd like to get that number down to a more normalized level of about 50,000 tons by year-end. As of today, we have sold forward almost 2 million tons or about 90% of our overall projected sales. We have sold to both domestic and export customers. All of our export -- pardon me, all of our domestic business this year is to repeat customers. We have sold coal at both fixed and indexed prices. And our fixed-price sales should provide us cash margins from our Elk Creek operation in roughly the $50 per ton range.

We will continue to explore ways in 2019 to increase overall production and sales. At our Elk Creek prep plant, we are looking at expanding the throughput capability levels and accelerating some new production we had slated for the 2020, '21 period. We are also considering adding some possible new High Vol A production at our Knox Creek Complex, both of which Mike will discuss shortly. We are not ready to pull the trigger on these projects at this time, but assuming we start a bit later in the year, most of the impact on our CapEx would be in 2020 forward.

I now want to spend a moment and discuss cash mining costs because I want to make sure that investors understand that there is a clear distinction between our cost structure at our 2 mine complexes as they analyze our results.

We ended 2018 with overall average cash cost of about $63 a ton. That broke out at $60 a ton average at Elk Creek and $127 per ton at our Berwind development mine. As mentioned, at Elk Creek we should had about 1.8 million tons production this year as we ramp up to, ultimately, roughly 2.3 million tons over the next 3 years. We expect cash mine costs at Elk in the mid-$60s per ton, up about 5% this year. About 75% -- pardon me, about 70% of that increase is due to higher sales-related cost based on higher sales realizations.

At Berwind, 2019 is again going to be another year of development mining. We expect production of about 250,000 tons this year in the thin 30-inch Pocahontas #3 seam with our overall mining-related cost at roughly $125 a ton. We'll reach the much thicker at 60-inch Pocahontas #4 seam in mid-2020. At full capacity, we expect production at Berwind at over 750,000 tons and cash margins near the $50 per ton level.

As Berwind ramps up and Elk Creek continues to produce as expected, we should reach close to a 3 million ton annual production rate in 2021. By 2022, '23, we plan to have production in the 4.5 million ton range. As we start generating meaningful cash flow -- free cash flow, and as we've said before, we would anticipate exploring with our board to start to return that cash to shareholders in the form of a recurring dividend. Ultimately, we anticipate being a coal company that is both growing as well as returning cash to shareholders.

I'd now like to turn to some of our forward views on the macro environment. We obviously like the conditions in this market and don't see either meaningful new capital or new production entering the U.S. met space anytime soon. As such, we continue to remain encouraged that the market still has legs well into the 2019, '20 period. A few signposts we are looking at: met coal spot prices are now at $213 a metric ton. The 2021 forward curve is up to $176 a ton. At the time of our last call in November, the 2021 curve was in the $160 level, so we are up about 10%. So while spot prices continue to be volatile, the forward curve continues to move higher.

Peabody recently noted in their view, industry CapEx is down over half from the peak. And we believe this lack of new investment is a big reason why the curve continues to flatten at higher levels. In addition, Chinese met coal production fell about 2% last year despite prices, generally, remaining strong all year. For 3 years in a row, Chinese met production has either been flat or down. We think it's safe to say that the government continues to push for supply-side reform, and again, that's another reason for the flattening of the forward curve at these higher levels.

M&A activity, especially international in the met space, continues to be one of the more active commodity areas. Rio Tinto's $4 billion sale of its producing Hail Creek and Kestrel mines was big news in '18, but we've also seen some under the radar met development projects in 2019. Gina Rinehart, who, as many of you may know, is one of Australia's richest women and the Chairman of Hancock Prospecting, recently put in a bid to buy the 80% of Riversdale's resources she doesn't already own in Canada. Riversdale is basically an unpermitted nonproducing reserve with a 4-million ton per year potential. That is a big number for a met mine that is not permitted nor producing. Indeed, if you look at it, that is almost our profile, and of course, we are in production and about a 2x valuation from where we are trading now for that particular metric.

Lastly, for those of you following trading metrics, I would note that we are now selling for about 1.5x book value and under 3x our consensus 2019 EBITDA. Given the twin black swans of 2018, with both the silo and the sandstone, I can understand that the market might be waiting to see some forward momentum for us to trade in a more normalized range with our peers. We hope that 2019 will be the year that we can demonstrate that momentum and then let both our cash flows and the market appropriately reflect that.

So with that, I'd now like to turn over the floor to Mike Bauersachs.

Thank you, Randy. The fourth quarter 2018 was very solid from an operating standpoint. Our mines continue to operate even while we were addressing the silo issue at Elk Creek. The ability to continue to run our mines has allowed us to keep our workforce intact. In fact, the productivities in our Elk Creek deep mines during January were some of the best we've experienced since inception. Included in the accompanying PowerPoint slide that illustrates the production build-out at Elk Creek. The high productivities and temporary capacity limitations of the plant have created unprecedented raw coal inventories at Elk Creek. We continued to be challenged by the size of the stockpiles, which were 330,000 raw tons at the end of 2018 and increased over 430,000 raw tons at the end of January 2019.

These inventories have caused us to incur additional handling costs and in multiple cases to lose unit shifts during the first 2 months of 2019. This issue should dissipate as we enter the second quarter, but it will take most of the year to fully work through the stockpiles.

Limited plant capacity has also caused us to miss multiple planned shipments to our customers. While we will continue to see some near-term impacts, we believe that we will be able to manage these challenges and meet our forward-issued guidance. It is a testament to our management team that we were able to quickly address the impacts of the silo failure at Elk Creek, and 3 weeks after the incident, resumed processing and shipping coal in December. While we have not been processing at 100% of our nameplate capacity, we were able to meet the most pressing needs of our customers.

I thought that it might be a good idea to provide some visuals to illustrate one infrastructure we have in place at the Elk Creek plant post the silo failure. Included in the PowerPoint is the picture of the processing facilities prior to the loss of the silo. This slide also contains picture of the internal failure of the cone on the day of the collapse.

The next slide is of the facilities today as well as a picture of where the permanent bypass system discharges into the plant. The following slide is of the temporary bypass, which was constructed primarily by a group of our deep mine employees at Elk Creek. This simple but effective bypass has served us well. Both the temporary and permanent bypass will remain in place until we have made all of the necessary upgrades to the silos. We are in the process of bolstering the cones in the 2 remaining silos from the bottom utilizing separate and dedicated foundation work. This should be complete in early April, at which time we project returning to 100% of our nameplate capacity of 700 raw tons per hour.

As we have previously relayed, we have insurance in place for both property damage and business interruption. To date, we have not collected insurance payments and have funded our extra expense, capital improvements and corrected measures out of cash flows. We remain firmly committed to the pursuit of a fair resolution in our outstanding agreement with our insurance company, who to date has chosen not to accept the claim or provide coverage. We strongly believe that it was impossible to predict any sort of defect or to determine exactly what caused the silo failure. We continue to make good progress in our development work at our Berwind mine. I've included a high-level mine map in our presentation to help illustrate the ongoing work at Berwind. The map depicts both the old works as well as the projections that lead to the area where we will slope up into the Pocahontas #4 seam. It also illustrates, in red, 2 horizontal long-hole drilling results. These penetrations have confirmed the coal is in place in areas that we intend to mine. Extensions of these holes also encountered the location and limitations of the sandstone washout that caused us to alter the path of our development mining last year. Both holes were successful in helping us gain confidence in what lies ahead and our goal of reaching the interesting slope location.

We have been fortunate to be able to sell Berwind low-volatile coal in 2019 for development mining at what we project to be a slight profit. While our development mining costs distort what would be our average costs, it is obviously preferable to generate cash while conducting what might have been capitalized mine development under less favorable circumstances.

Note that we have also added a second section to this mine. It is currently operating at one shift per day, but we anticipate it to be fully functional in the second quarter. The capital cost to bring on this additional production is fairly minor due to the existing infrastructure and should help us lower our operating costs as the year progresses. This additional production will also feed the Knox Creek plant, where we have available capacity. This active section will also help us to smooth out our coal sales to customers, while we are migrating our mining to the Pocahontas #4 seam.

Since it will likely take a large portion of the year to work through our stockpiles at Elk Creek, our near-term focus on increasing production is continuing to develop Berwind as well as evaluating the Jawbone high-volatile A reserve at our Knox Creek property, which we estimate could produce approximately 500,000 tons per year from 2 sections. One of the ways to reach the high-quality Jawbone is to utilize existing infrastructure in the idle Tiller seam mine that lies slightly below the Jawbone seam. This could significantly reduce the capital cost of the Jawbone seam mine. We have recently viewed the old works in the Tiller seam mine and the conditions are promising for a restart.

While it is premature to announce that we've committed to develop the Jawbone mine, our review of the economics to date has been positive. We will continue to update you on our progress relating to this near-term opportunity as the year advances.

During 2018, we determined that the biggest and most material miss in our original projections and our go-forward mine plans was that our surface mine will make up a smaller portion of our future production. In particular, we will likely only have 1 surface mine, not 2. With that being said, our existing surface mine has steadily improved its productivities and consistency throughout 2018.

I have included a slide that highlights these trends. This mine should operate at approximately 350,000 tons per year. We look forward to operating what will likely be our lowest cost form of production for a decade or more to come. The only other major variance is the delay in the development of the Berwind Pocahontas #4 seam, which is now back on track. We are working hard to find a way to bridge the gap created by less prolific surface mining. The Jawbone opportunity mentioned before is one of the additional organic developments that we are reviewing to reach our previously projected 4.5 million ton per year run rate.

One additional opportunity that we are also reviewing is not in the form of a new coal mine, but instead upgrading the Elk Creek plant to increase the raw feed rate. This would potentially allow us to increase production by 500,000 tons and it would likely push forward planned production from new mines. The analysis of this opportunity is ongoing, and we will continue to update this capital project as the year progresses.

Recently, we've seen an announcement from one of our competitors to develop a new long-hole mine. We also saw an announcement from an additional competitor relative to a potential new mine. Both are or will be high-volatile A mines. As we look at our plants, we continue to be encouraged by the low-volatile marketplace and the future needs of our current and prospective customers for our Berwind coal. The fact that there are no new announcements in the U.S. for substantial new low-volatile production, continues to reinforce the value of our Berwind coal in the domestic and international marketplace.

Overall, we believe the lack of available capital and the decisions by many of our competitors to issue dividends, substantial special dividends as well as buy back stock will keep supply in check.

We enter 2019 with only around 100,000 tons being carried over from 2018. This carryover business on average is priced below what we have sold in 2019 creating a small negative variance to our realizations for the first quarter. We've incurred no cover or related claims associated with the silo issues from our customers. We have continued to enhance our sales mix with domestic customers in the fourth quarter of 2018 and early 2019.

We enter 2019 with one of the strongest and most diverse customer basis in the sector. We currently have business in place with 6 domestic customers and have index business with 4 customers for coal that is designated for export. While the sales mix has changed, we have retained business with all of our domestic customers for 2019. I've illustrated these commitments in the accompanying PowerPoint. Note that we will likely ship a significant larger number of tons versus production in 2019 due to existing stockpiles.

2019, from a marketing perspective, will be the year where our coal will meet its realization potential and becomes fully aligned with the quality we ship. The sheer number of return customers and new business is a testament to how well the coke plants that use our coal like the quality in their plants. While our capital deployment in 2018 exceeded both our budget and guidance, the multiple items that we focused on will serve us well going forward.

We have added a set of dual plate presses to our Elk Creek infrastructure, a picture of one of plate presses is included in the PowerPoint. This equipment will allow us to better and more fully utilize our impoundment as well as extend the life of the current refuse and impoundment footprint to somewhere around 20 years. One of the most critical capital investments in 2018 was to pave our main haul road at Elk Creek. With the negative weather impacts incurred during the winter months, paved haul road allowed us to keep feet on the plant and is one of the key reasons why we had a positive fourth quarter.

Let me also note that the largest negative variance to our projected capital spend was associated with our Berwind development in the form of capitalized mining costs. Most of this was incurred as we explored and worked our way around the previously disclosed sandstone washout.

Another notable capital expenditure was the purchase of mobile roof supports for our Eagle mine at Elk Creek. This equipment is a unique design that was developed through a partnership with J.H. Fletcher. It allows for supports to operate in a base range of 6 to 18 feet. It also accommodates add-on components to extend the already elevated reach of 18 feet to a total of 21 feet. I have attached a picture of one section of the equipment. These supports are also a byproduct of our commitment to safety. While completing what should be highly productive secondary recovery, let me also note that in a competitive market for quality miners, the deployment of safety-driven capital assisted with employee retention.

Our projected CapEx of between $35 million and $40 million for 2019 contains the funds for advancing Berwind development; potential stockpile improvements at Elk Creek, which includes finishing the previously deferred third clean coal stockpile too, it also includes the spending for the silo upgrades and bypass system. This estimate includes only nominal work associated with the Jawbone seam, which should be in addition to the aforementioned range. We are poised to generate substantial free cash flow during 2019. Moreover, we anticipate a substantial jump in production in sales during 2020.

I would like to conclude my remarks with a focus on employees and safety. Thankfully, all employees were safe in conjunction with the silo failure. While other companies might have made a decision to reactivate the remaining silos at Elk Creek sooner, it was never really a consideration for us. Our employees and their safety are just too important. We're going to great lengths to make safety-enhancing improvements to the remaining silos as well as ensure their use for the life of our mines. In conjunction with all of this, we also took great care in safely demolishing the silo as well as found ways to continue to produce coal, while the plant was idle. In return, our employees have been very loyal and understanding. Turnover has been very low even as we have had to eliminate some unit shifts to control stockpiles.

On that note, I will turn things over to Mike Windisch for some final comments relative to our fourth quarter and full year 2018 financial performance.

Thank you, Mike. I'd like to refocus the discussion on our fourth quarter and full year results. During the 3-month period, we reported net income of $3.4 million or $0.08 per share and adjusted EBITDA of approximately $7 million. Both represents decreases sequentially over the third quarter, but obviously a huge increase year-over-year.

For the full year, 2018 net income totaled $25.1 million or $0.62 per share and adjusted EBITDA was just over $42 million. Total revenues during the fourth quarter were $44.2 million, down significantly from the prior quarter. Our margins on company produced coal, however, did improve in the quarter to $28 per ton, up from $25 per ton in preceding 3 months.

For the full year, revenue totaled $228 million, which was a 273% increase over 2017 levels. Our cash mine costs saw a considerable improvement during 2018 as well, dropping from $73 per ton in the prior year to $63 per ton in 2018. Including significant capitalized development costs at our Berwind mine, CapEx totaled approximately $8.3 million during the quarter and $48.1 million for the full year. This includes some incremental CapEx at Elk Creek relating to our silo failure and related workaround.

During the fourth quarter, we greatly improved our working capital flexibility with the closing of a combined $40 million 2-year amortizing equipment loan and a 3-year ABL with KeyBank. We had only $9.6 million of outstanding borrowings with an additional $33.4 million of available liquidity at year-end.

This now concludes management's prepared remarks. At this time, I'd like to open up the line for any questions you might have on our 2018 results or outlook for 2019.

(Operator Instructions) Our first question comes from Jeremy Sussman of Clarksons.

Randy, I guess, starting here, you noted that as production increases, you anticipate exploring with the board the possibility of returning cash to shareholders in the form of a recurring dividend. Can you give us a bit more color on maybe what you're thinking here?

Sure. So last year, of course, we did not have free cash flow. We basically were slightly below. This year, we anticipate our first year of free cash flow and that sort of begins a trend as we project forward to have pretty meaningful cash flow as we go out between now and, let's call it, 2021, '22. So I think, once we get to a point where we're comfortable on having a fairly steady recurring level of free cash flow, we have always said from the very beginning that we would explore dividend policy and we continue to do so.

That makes sense. And maybe secondly, Mike, this may be a question for you. I guess, 2 projects that you talked about caught my eye. First, the potential to add 0.5 million tons or so of high-vol A at Knox Creek and then secondly, you noted that you're looking at ways to add 0.5 million tons of -- secondly, you noted you're looking at ways to add about 0.5 million tons or so of throughput at the Elk Creek prep plant, which could allow you to accelerate some production. So I guess, could you expand on these opportunities? And maybe, in particular, what are you seeing that could make them attractive enough to pull the trigger on later this year?

Yes. Sure, Jeremy. I think since we have Chris Blanchard with us, I'm going to go ahead and let Chris address those. He's a little bit closer to it than even I am.

Jeremy, starting at Elk Creek, we have 2 permitted mines that are high-vol A quality. They are centered in some of the seams that we're currently mining, but there the quality is far superior and they're fully permitted that we could bring online. They were in our mid-intermediate term plans anyway, but we could accelerate them given extra capacity at the Elk Creek plant. They're both located in pretty close proximity and belt to the plants, so they've got a good cost profile even though they're slightly thinner. But obviously, we need to upgrade the plant throughput a little bit.

When the plant was designed and built, there is some excess capacity on certain ones of the circuit, some of the fine coal circuits. So we don't have to do a full build on the plant. We have to do -- some of the coarse coal and some of the conveyors have to be beefed up a little bit, so it appears to be relatively modest capital for the improvements at Elk Creek and the increased throughput.

Switching to Knox Creek, we just have -- we have a permitted coal mine. We've explored it, reactivated the permit in the deep mine and explored it and found the conditions underground to be really quite good for a mine that had been idle for 6 or 8 years. So the incremental CapEx to develop the -- to develop into the Jawbone seam at Knox Creek, which is a high-vol A product, is also relatively minor compared to having to sink shafts and slopes or do major development work there. We think we could bring on 0.5 million tons there over 18, 24 months and the labor in that area is also pretty conducive to bringing on that extra production, the labor availability that is.

Yes, just to add a couple of things to the Knox Creek thing, having the availability to access an existing prep plant with all the infrastructure there is also a big thing that I think probably makes that Knox Creek project even more compelling maybe even in some of the stuff at Elk Creek. And having a high-vol A coal on the Norfolk Southern, as you guys know, is a big plus because there's more low-vol really there than high-vol. So we're really excited about that opportunity.

And our next question comes from Curt Woodworth of Crédit Suisse.

First question just on Berwind for 2020. So if you get into Pocahontas #4, say, midyear, what's the timing from then to ramp up to the 750,000? And then just from a modeling basis for 2020, how should we be thinking about costs? I assume, it'd be $125 at the start of the year and then trending down, but any color on that would be appreciated.

You talk about production, go ahead.

So this is Chris. At Berwind, obviously, we should reach midyear development with one of our operating sections. It will start branching out and driving mains for the other section, and that will be -- the remaining 6 months of 2020 will probably be -- I, sort of, hesitate to call it, development mining, but it will be setting up to put the mine in full production mode for 2 sections. We'll move to second section in 2021, and we'll have 2 sections operating full year in '21. We should see the production ramp up, but we're really not expecting the full run rate until the latter half of '21, and so we won't get full production until 2022 out of the Berwind mine with 2 sections running full year at full production in the Pocahontas #4 seam.

Okay. So would you say that cash costs in 2020 should be somewhat similar to 2019 or slightly low as you enter the first section?

The first half of the year should be similar to 2019. As we get into the second half of the year, we should see the improved cash cost on the number one section. So for the second half of the year, I would expect us to be halfway between '19 and full run rate production for the blend and then '21 and on when we have them both in there, we should be towards our long-range cost guidance.

Okay. And then, Mike, on Jawbone, what is the reserve base there? And what would you estimate cash cost to be?

Yes. The reserve base, I think, is somewhere around 5 million tons or so -- 5 million to 6 million tons recoverable. The -- when we -- I guess, when we think about the cost structures, which -- it's going to be a little bit thinner than some of our Elk Creek coals, which, obviously, are exceptional.

I think you've probably got cost in the high 70s and low 80s depending on the fluctuating clean ton per foot. So still very compelling for a high-vol A coal, of which -- it will -- until we put the Glen Alum Tunnel seam in at Elk Creek, it would be our best product that we have. So real excited about it because really slightly higher costs are offset by what we believe will be substantially higher revenues.

Okay. That's helpful. And then just last question on the -- in the portion of your business that's priced on an index basis, can you talk about how that map works? So is the index based on like U.S. East Coast high-vol A number and then discounted back? Or how should we think about incorporating that into our models?

Yes, we've got some different pieces of business, but the right way really to look at it for us is to kind of take a -- look at a high-vol A number on the East Coast and B and really kind of strike a number, sort of, in the middle of that. Most of our coals -- although we do -- we sell some coal off -- purely off the high-vol A minus a certain amount and some B plus, but really when you take it and you put it all together, it's really -- it's a high-vol A-B coal. So just strike it right down in the middle, I think, it would be pretty good for your projections.

And would that be based on like a quarterly lag or a monthly lag in terms of how the price resets?

Yes, it's really the month that it ships for the most part. So we, kind of, come back and calculate it. And they're not all exactly the same, but that's really kind of how it works.

And our next question comes from Mark Levin of Seaport Global.

Great. I have 2 questions. One more modeling and then the other bigger picture. So on the modeling one, what type of rail and port rate should we assume in 2019, so the rail plus the port charge?

Yes. It's -- we've seen a little bit of escalation in our port rate. When things get pretty good, the numbers are fairly substantial. So I mean, you're going to see, depending on how the pricing goes from the mid to high $30s to $40 with both those numbers. So it's -- obviously, it moves around a lot based on the market, but we've seen just a little bit of -- especially port charges seen a little bit of escalation there from what we saw in '18.

And Mike, just putting a finer point on that, how much do you think that would be up year-over-year in '19 versus '18?

Yes, a couple dollars for sure on the port stuff, and the rail has been fairly consistent. So I would say that even on the rail, maybe it's $3 or $4 up from where we were all in.

All in. Got it. Okay, great. And then, this is bigger picture. So I think you mentioned -- or Randy, you may have mentioned getting to 4.5 million tons in that kind of '22 to 2023 time frame, so 2 more specific questions around that. When you think about the 4.5 million tons of potential production, can you maybe put them into different buckets, whether it's Elk Creek, Berwind, Knox, just the different buckets? And then as a corollary, when you think about what the overall cash cost profile might look like of the enterprise, when you do get to 4.5 million tons? And I realized there's a lot that goes into that, whether it's sales-related cost or inflation, all sorts of things to try to make it costs, call, several years out. I understand the complexity of that. But just generally speaking, where the production is going to come from and what the ultimate cost profile could look like at a 4.5 million ton production rate?

Yes. I think it's a little bit of a moving target as I'm sure you could appreciate looking several years out, but I think if you just, sort of, took large buckets at Elk, we're probably going to be 2.5 million to 2.75 million. We've got our RAM Mine up in Pennsylvania, which is about a 300,000-ton operation. Knox Creek, let's call it, 0.5 million tons there. Berwind will be just under 1 million tons. So that gets you in, sort of, the 4 million, 4.5 million ton range. And I think in terms of our cost structure, I think, probably overall particularly with some of the new mines we're talking about, we would probably be in the, sort of, low mid-70s, somewhere in that range.

And our next question comes from Michael Dudas of Vertical Research.

First question, I mean, following up on Mark's -- on the outlook longer term, what type of return requirements do you guys look at relative to these capital investments? And as you look at the balance sheet and look at, obviously, the potential of free cash balancing the growth and the investment, what the capital structure of the company should look like optimally? Or how would you like to see it over the next couple of years as you're meeting this balance between growth in CapEx and returning cash to shareholders, mindful of the stock that's trending it, as you mentioned something times EBITDA?

So Mike, this is Randy. Let me kind of start in reverse order. So I think our mantra has always been to be, sort of, a no-debt or relatively low-debt operation, and we don't see any reason to discontinue that outlook. All the things that we're talking about are pretty much organic projects. We are not talking about any type of large acquisitions. It's really taking reserves that we've now got and then essentially just developing those and frankly trying to develop them in a fairly low-cost manner. So I don't think that you would see us layering on a lot of debt into our capital structure. We've got a revolver now, which, of course, is very helpful for cash management, but we don't anticipate putting on a lot of new term debt or anything of that nature to kind of ramp up to do deals, that -- if that addresses where you're coming from there. And the first part of your question, again, was...

Yes, he was asking about return...

On the return calculation. So I think we don't approach it necessarily with a strict mathematic that we're going to try and get up a 15%, 20% return, although any of our projects would, frankly, exceed that by a reasonably substantial measure. A lot of what we're trying to do is just, sort of, strategically position ourselves, so that we got a pretty good blend out there of various products that we think have got high-margin potential.

I'm just going to add. I mean, in the coal business, with all the potential for revenues fluctuating key component, of course, is low cost. And when we run the numbers, absolutely, a net present value-type calculation that has discount rates of 20% plus just because of the risk.

I think to just echo that, I mean, again, if you go back to our, sort of, original mantra, if we were trying to be both low cost and a little or no debt, and we, sort of, figured that we were in that position, we will be able to do ride out a few storms and certainly be able to be comfortable in terms of whatever market conditions that otherwise somewhat volatile commodity might bring us.

And Randy, just maybe a follow-up, how your major shareholder feel about how things are working in the plant, et cetera and their long-term visibility potentially in the investment?

Well, our largest shareholder is a private equity group called Yorktown, which, of course, has been with us since the inception. They are frankly very happy about where we are. Their long-term expectations are to approach us a core portfolio investment, where they have essentially modeled their outlook on a couple of other deals that they have done, which, frankly, they've held for over 30 years, which are very nice low debt, high cash flow, producing entities where you essentially have the same strategic plan we have, which is to return shareholder cash in the good years and be able to be flexible in the bad years, perhaps even consider buying back some stock. So our long-term plans are to have a very healthy cash flow producing machine and that gives us a lot of optionality.

Terrific. Just one final quick one from me for Mike Bauersachs. Mike, as you look into '19, maybe if you can share some views on the acquisition market, maybe on what you're looking at, your small deals you'd like to swap and especially enhance what you guys have around your reserve and your deposits? But also, like how you see given the cost pressures and the lack of growth, et cetera, how -- is there going to be more competition or some more decisions in the marketplace, especially in Central App on the met coal front that could maybe enhance Ramaco's opportunities on a relative basis?

Yes, we're constantly looking at things next door. We kind of -- we like smaller deals. We like to do things that a lot of people don't. We'll buy assets that are not accretive. And we believe that we're pretty good at deploying the capital in geologically advantage properties. And indeed, there are some properties that we're looking at. We've got the potential to lease some additional reserves. They're very close to our existing properties. These are things from a risk standpoint that are very low-risk wise. We've got some infrastructure projects that -- or some infrastructure opportunities that are kind of close to some of our properties that we think might be cost enhancement-type acquisitions. But as far as larger transactions, a lot of stuff that we see, we just hate liabilities, Mike. As you know, it's been, sort of, one of my focuses in the past. And large mergers, they have large amounts of ARL liabilities and those kinds of things are not the things we're interested in. We, over time, want to keep the balance sheet pretty clean because we think we win over time with that, sort of, balance sheet. So as far as a larger picture, nobody has a lot of cash to do deals with. Nobody's stock currency is such that you feel like there could be some, sort of, stock deals. So as far as the domestic side, it's pretty tough to envision anything substantial happening from at least my view.

I will say, Mike, to go back, again, to sort of a macro, if you look at what we're trying to do organically, which admittedly has kind of hit a lot of singles, maybe couple of doubles, but certainly not trying to go for homers. There's not a lot of people out there we see that are trying to essentially double the footprint of their company in the next couple of years. So at least in that respect, I think, we somewhat stand out from the crowd.

And do it from free cash flow, which is the right way to do it.

I think keeping that balance sheet clean is a very good priority.

And our next question comes from Lucas Pipes of B. Riley FBR.

I wanted to ask if you could remind us of the capital intensity of the various growth projects kind of what would be required this year, next year, the year after that? And then any updated thoughts on maintenance capital as well?

Yes, I'm going to go ahead and defer to Chris. I can say that we're continuing to evaluate a couple of these things. So there's -- some of this is, sort of, a moving target, but for the most part, a lot of the mining-type stuff is more or less equipment and/or driving slopes and accessing the existing infrastructure, for example, at Knox Creek. But with that being said, I'll kind of turn this over to Chris and let him give you an idea, and he can also talk about, sort of, maintenance CapEx overall for the company and what our baseline numbers are.

Yes. So for the growth CapEx that we're considering, it might be a little bit premature to throw a number out to you guys on those. But as far as what we're projecting in, sort of, our ongoing plan or a public numbers, we've got about $10 million of capital that we're going to deploy at our Elk Creek plant, which is both related to our third clean coal tube and handling facility that has to do with completing the silo repairs and the workarounds and some other small projects around the plant, but that's about between 1/3 and 1/4 of the total spend for Ramaco for '19 will be the Elk Creek plant. And then on an ongoing basis, our mines are still relatively new, most of our underground mines are continuing to develop, so we're advancing belt, advancing structure, buying belt -- drive that type of maintenance CapEx. And as we get a little bit more mature, we'll have continuous minor rebuilt and that, sort of, thing coming.

And if you want to use a rough number, somewhere in the $6 to $6.50 per clean ton of production number for maintenance CapEx ongoing at our underground operations and less at our surface and our plant. And then just rough numbers, really, really broad brush, some of these expansion projects, that we're talking about, were in the range of $5 million to $10 million incremental a piece.

Over several years?

Yes, over several years. So that keeps me from hemming myself in too tightly, but gives you a little bit of a flavor for the size projects we're looking at.

That's very helpful. And Randy, Mike, maybe a higher level industry question. What you would expect kind of U.S. met coal production as a whole in the aggregate to shake out in 2019 versus 2018?

I think from a production standpoint, overall, I think we had about, what, 750 million tons produced last year just in both thermal and met markets. I see that we'll still be trying to produce as strongly as we can next year or rather this year given the state of the markets.

So I would see a slight incremental jump. I think the governor on all that production is, frankly, capacity. I think most of the mines that we serve, they seem to be running at fairly full out. And I think a couple of the development projects, particularly the larger ones like Arch's and Warrior's are probably not going to hit here for a -- maybe a couple of years. So I think, this year, I would say a modest incremental bump.

(Operator Instructions) Our next question comes from Nathan Martin of Seaport Global.

Just a quick modeling-related question. I mean, in the past, you guys have given guidance on purchase coal, I guess, you did about 427,000 tons in 2018. Any thoughts on purchase coal in 2019? Obviously, you do have the unprecedented stockpiles there. So just any color you could give us there would be great.

Yes, we have -- we really turned our Knox Creek plant to focus on our internal production, although we are purchasing some coal, but this year, I envision that number going down to somewhere around 100,000 from one mine that we're focused on. I will say that as we get past this year and we begin focusing on some other international markets from a future standpoint, I think, that number could grow. But from a modeling standpoint this year, I would show about 100,000 tons.

And that, by the way, is probably going to be low-vol.

And I am not showing any further questions at this time. I would now like to turn the call back over to Randy Atkins for any further remarks.

Once again, we just appreciate everybody being on the line this morning and look forward to catching up with you very shortly as we talk about our Q1 results in May. Thank you very much.

Thank you, ladies and gentlemen. Thank you for participating in today's conference. This does conclude today's program, and you may all disconnect. Everyone have a wonderful day.