Liberty Energy Inc (LBRT) 2017 Q4 法說會逐字稿

Good morning, and welcome to the Liberty Oilfield Services 2017 Fourth Quarter and Full Year Earnings Conference Call. (Operator Instructions) Please note, this event is being recorded.

Some of our comments today may include forward-looking statements reflecting the company's view about future prospects, revenues, expenses or profits. These matters involve risks and uncertainties that could cause actual results to differ materially from our forward-looking statements. These statements reflect the company's beliefs, based on current conditions that are subject to certain risks and uncertainties that are detailed in the company's earnings release and other public filings.

Our comments today also include non-GAAP financial and operational measures. These non-GAAP measures, including EBITDA, adjusted EBITDA and return on capital employed are not a substitute for GAAP measures and may not be comparable to similar measures of other companies. A reconciliation of net income-to-EBITDA and adjusted EBITDA and the calculation of return on capital employed, as discussed on this call, are presented in the company's earnings release, which is available on its website.

I would now like to turn the conference over to Liberty's CEO, Chris Wright. Please go ahead.

Good morning, everyone. I want to welcome our new investors to the Liberty family and thank you for joining us on our first earnings call as a public company. We're proud of our fourth quarter and full year 2017 results.

Over the course of 2017, we nearly doubled our fleet count, expanded our presence in the Permian, entered the Eagle Ford and added over 1,000 new team members to the Liberty family.

We did all this while generating industry-leading returns and earnings per fleet. Our full year revenue totaled $1.5 billion with net income of $169 million.

Full year adjusted EBITDA was $281 million or $18.6 million of adjusted EBITDA per average active frac fleet. In the fourth quarter, our revenue was $449 million, generating net income of $58 million. Adjusted EBITDA totaled $92 million, or $20.2 million per average active frac fleet on an annualized basis.

We are a returns-driven organic growth company, focused on compounding shareholder value by reinvesting cash flow at high rates of return. For the full year 2017, we generated pretax return on capital employed of 35%. From Liberty's inception, we've had a vision to build a different type of company, one with a foundation rooted in technology, committed to long-term partnerships with our customers and suppliers.

We have built this company from day 1 with the overriding idea that, if you bring together the right people and provide a culture that fosters innovation and efficiency, they will achieve outstanding results.

Technology and innovation are embedded in the DNA of Liberty. Technology has enabled us to build long-term customer partnerships and has been a key driver in generating our differential financial results.

We are firm believers that the best ideas are the ones that solve specific problems, from frac design optimization to maximizing operational throughput, minimizing maintenance cost and reducing environmental impact. We are driven to reduce our customers' cost to produce a barrel of oil. We've focused significant resources to develop our extensive geologic and engineering database of unconventional wells. We're utilizing this database as the starting point for our in-depth proprietary analysis to recommend improvements in completion design in the Permian Basin, as we've done for years in the Rockies. Simply put, we want to help our partners maximize their returns. We're excited as we move into 2018, the frac market remains very strong. Improved commodity prices and increased drilling activity are providing a positive industry backdrop. The demand for frac services continues to outstrip supply. We believe this trend will persist through 2018, supporting modest upward pricing potential. We expect that 2018 will be a year for the industry to focus on quality of execution, which is what Liberty does best.

More so then I earn, we believe that human constraints in the workforce will continue to regulate the pace of the industry expansion. Having retained our people through the last downturn, we believe we have uniquely experienced workforce and the ability to attract new team members more easily than others.

Our fleets are long-lived assets as are the people who join the Liberty family to operate them. Keeping the positive market backdrop, our record of performance and excellent customer relationships, Liberty continues to experience significant excess demand for our services.

During the first quarter, we deployed our 20th and 21st fleets under dedicated arrangements with existing customers and as previously announced, plan to deploy our 22nd fleet on a dedicated basis at the end of the second quarter.

We take a long-term, through-cycled approach when we decide to commission a new fleet. We always want to ensure demand for Liberty significantly outstrips supply. In deciding when and where to add assets, we evaluate where we believe we are in the cycle, our existing or potential customers and their long-term plans and the ability to staff a fleet with the quality of people we demand. Based on these criteria, we have placed orders for 2 additional newbuild fleets to satisfy commitments we've made to customers. We expect these fleets will be delivered in the third and fourth quarters of 2018. With this newly committed construction, we expect to exit 2018 with 24 fleets in service.

I will now hand over the call to Michael Stock, our CFO, to discuss our financial results.

Good morning. We're extremely pleased with our 2017 results and believe we have put in place an excellent platform to support our growth going forward. We grew our fleet count from 10 active fleets to 19 over the course of 2017, while increasing operational throughput and maintaining a strong record of safe operations.

For this call, we will discuss our net income and EPS on a pro forma basis. As far as IPO, Liberty's predecessor companies were not subject to U.S. Federal Income Tax at the entity level and as such, our net income does not include income tax expense.

As for our former adjustments, they're only to reflect what our income tax expense and subsequent net income would have been had we been a public company for all of 2017.

The fully diluted earnings per share calculation reflects the impact of additional tax on net income and shares outstanding issuing the conversion of all of Liberty's Class B common stock into Class A common stock.

For 2017, we generated pro forma net income of $130 million and fully diluted earnings per share of $0.88. Revenue for the year was $1.5 billion, and adjusted EBITDA totaled $291 million.

Adjusted EBITDA through our average active fleet and return on capital employed, as previously described by Chris, are 2 key metrics we use to evaluate our performance.

For 2017, we generated adjusted EBITDA through average active fleet of $18.6 million and return on capital employed of 35%.

Revenue for the fourth quarter grew to $449 million, a 2% increase over the third quarter revenue of $442 million.

For the fourth quarter, we generated pro forma net income of $45 million and fully diluted earnings per share of $0.30, compared to third quarter pro forma net income of $49 million and fully diluted earnings per share of $0.33.

Adjusted EBITDA for the fourth quarter totaled $92 million or $20.2 million of annualized EBITDA per active fleet, compared to $22.3 million per fleet in the third quarter.

In the fourth quarter, we experienced an impact to our operations due to holidays and some customers having exhausted their full year budgets earlier than expected. While curtailed activity did impact our fourth quarter earnings, we believe disciplined spending across the upstream space will serve to provide greater balance to commodity prices and extend the current cycle.

General and administrative expense, excluding fleet activation costs totaled $17.6 million for the quarter or 3.9% of revenue.

We expect general and administration expense excluding noncash share-based compensation to decline slightly as a percent of revenue in 2018, as higher revenues offset the increased G&A costs associated with being a public company.

We ended the year with a cash balance of $16 million and total debt $196 million. On January 17, we completed our initial public offering. We utilized a portion of the proceeds to complete the pay down our ABL facility and a portion of our term debt.

Pro forma for the offering at year-end, we had $118 million of cash and $107 million of debt.

In the first quarter of 2018, we experienced some extreme and unusual weather across all 5 of our operating basins, reducing throughput and in some cases, temporarily shutting down operations.

While these events will impact our first quarter earnings, we expect to see high single-digit growth in revenue and EBITDA of the fourth quarter. While our first quarter earnings may be slightly below our expectations, the strong demand for our frac services leaves us confident in our expectations for the full year.

As we look to 2018, we will continue to emphasize at disciplined organic growth plan. We expect to deploy a total of 5 fleets in 2018, consisting of 4 newbuild Liberty acquired fleets and our final refurbished fleet.

We are upgrading 5 of that fleets to meet customer demand for high-pressure work in the Permian and Eagle Ford.

We're constantly working to increase throughput. Towards that end, we're building capacity to include pump down services for all Liberty frac fleets and build ancillary pumping services business in our 3 major basins to perform pre frac services, including toe preps, acidizing and other services previously handled by third parties.

Our fleets are designed with the lowest total cost of ownership philosophy. Combined with our rigorous preventive maintenance program, we believe we have one of the lowest maintenance CapEx expenditures on a per fleet basis. Our maintenance CapEx budget for 2018 is $55 million or $2.5 million per fleet.

For 2018, our CapEx budget is approximately $310 million, which includes $20 million of carryover from the fourth quarter and the additional horsepower to provide ancillary services previously discussed.

We expect our 2018 capital budget to be completely funded by cash flows from operations. We expect to end 2018 with 1.2 million horsepower, consisting of 19 standard 40,000-horsepower fleets, 5 50,000-horsepower high-pressure fleets, 120,000 of dedicated pump-down capacity attached to these fleets and approximately 40,000 of horsepower supporting ancillary pumping service line.

With that, I would turn the call back to Chris, before we open for Q&A.

Thank you, Michael. We believe that 2018 will be a year defined by execution. We look forward to strengthening our customer partnerships and working together to increase operational efficiency and throughput, lowering the cost to produce a barrel of oil.

I want to thank our customers, suppliers and the whole team at Liberty who work tirelessly to advance the shale revolution that is changing the world's energy landscape. Thanks for joining us. I will turn it back to the operator.

(Operator Instructions) The first question comes from Sean Meakim of JPMorgan.

Chris, as you'd increase your mix in the Permian here going forward, so that you could just maybe elaborate a bit on how the mix changes there thinking about stages per fleet, revenue per stake metrics, just the variance across the DJ, at market versus the Permian? And how that should influence first half of per fleet, as you move through 2018?

Yes, you bet, Sean. I mean, in the Permian we -- since the reservoirs are thicker, the average size pounds of sand in a frac stage is bigger than it is in the DJ or in the Barkin. But obviously we get paid more for that. But we get paid even more on a per pound basis because the efficiency or average throughput in the Permian is still lower than it is in the Rockies, but on a gross profit per fleet basis, in the rearview mirror, recent rearview, I'd say they're about equal. I'd say Permian is definitely moving ahead of the other basins. It'll be slightly more gross profit or EBITDA per fleet than the other basins, but they're reasonably aligned today.

Right. So then your expectation would be that even if the efficiency mix is deteriorating, and does it necessarily have a negative impact on gross profit per fleet?

Yes, right now the pricing offset that. But we're actually seeing -- just recently some meaningful improvements in efficiency and throughput in the Permian. We like a lot what we see in the Permian.

Got it. Okay, and then just thinking about the supply chain, could you maybe just give us little more detail of how much loss work is a drag on margin in the first quarter? Are you paying top line in margin? And then just any other potential bottlenecks that you think are worth pointing out here?

Yes. We mentioned weather and some of the struggles we and really the industry had in Q1. And it did impact throughput and if you get less done it impacts to margins as well. But I'll let Ron speak to that, but I'm pretty proud about how our supply chain has handled the struggles in Q1.

Yes, Sean, certainly you've heard the challenges at the -- some of the industries talked about, specifically, around the railroad and the ability to move sand there. We've been fortunate to avoid any issues as a result of that, our supply chain team worked pretty hard to make sure that we had not only primary supply in basin but also secondary and tertiary opportunities there.

A big thanks to our partners on the supply side, who worked hard with us to make sure that we were able to keep things up and running. So very, very proud to say that, to this point in time, we have not yet experienced a shutdown as a result of sand issues.

Got it, that's very helpful. Maybe just play it a little bit differently, as you think about -- there's a number of factors driving your gross profit to be in the fourth quarter and the dating process as well, but just thinking about January versus March, could you maybe give us a sense of the trajectory into a quarter? And that could be helpful.

You bet, Sean. Yes, January for us was a most impacted month. Some extremely cold weather, shut down roads where you couldn't move anything, fuel or chemicals, in addition to sand. And just -- and problems with water supply, which are not handled by us and wire line from the extreme weather. So January was meaningfully impacted, February, maybe, a little less lower or similar, and March, definitely, definitely better. Yes, I would say we're at the highest throughput of the quarter right now.

The next question comes from Connor Lynagh of Morgan Stanley.

Just wondering if we could dial in a little more on just how much we should expect EBITDA per fleet to decline in the first quarter? And if we'd look back at the third quarter where you are doing $22-or-so million of EBITDA per fleet, is that a reasonable expectation for midyear this year, can you do better? With pricing higher and the efficiency moving higher in the Permian? Just what are your expectations on that front?

Yes, Connor. Yes, certainly, we expect to see exceed those numbers this year, this summer, wherever, in that timeframe where we don't have other impediments, pricing is improved, our team, our efficiency. They were able to deliver, has improved since then.

In Q4, we talked about exhaustion of budgets and holidays and things just less, less days on location moving sand and fluid through our pipes. Q1, we've had some similar struggle mostly for different reasons, although weather's an overlapping reason.

But I would say our expectation as we said in our press release for Q1 is not a decline in EBITDA per frac fleet from Q4. It looks to be roughly flat, similar growth in fleet to revenue in EBITDA. But yes, both of them are impacted versus the cleaner operating environment we had in Q3. But again, the macro backdrop for us today is better than what it was in Q3.

Yes, yes, absolutely. I was interested, the comment you made on the Permian, making steps up in efficiency, do you think that's a market. A wide comment or is that due to thing that you guys in particular are doing with your customers there? What are you attributed to?

Wide -- we should have more months to see how this unfold, but I would say it's probably more a Liberty-specific thing, in the Permian itself. It's just it's very busy ancillary services are stretched, and there's some obvious struggles you are going to have when is business is growing that fast.

If you're looked at the Barkin in 2013 or 2012, you would see the same stresses on a system before infrastructure and available services catch up with activity level.

But it comes down to the right partners and customers, and the right partners in supply chain and logistics, and just learn it how to deal with the different challenges in the different basins.

Got it. And just in terms of where you see it going from here, I mean, how much more efficient can you be in that market? Or just broadly across the U.S.? Is it 5%? Is it 30%? Just high-level, how bigger is the opportunity that's out there?

I'd say you probably bounded it nicely. Certainly, we've got more than a 5% upside in what we can get done with efficiency. I mean, I would say it's meaningful, it's meaningful. 30% is large, that may be several years down the road. But again, I would say the gains we expect to get are somewhere in that wide range. There's a number of initiatives, and you've heard of some of them in the press release. Pump-down operations, to get the wire line guns down between stages, we do that in some of our fleets but not in all of our fleets and in a stress market, we're seeing that hold us back. So we've decided, we don't want that to hold us back, so we're taking pump-down operations across all of our fleets in-house. That will help drive efficiency and reduce problems. Same thing for toe preps and acidize, other things that frac pads, so that they are on schedule to move it. We're rolling out this year an automated pressure testing system that not only does a better job pressure testing, but it's faster and more efficient. So there's -- I could go on and on. We have a number of operations, as you know that's the culture of our company just to figure out together with our customers how to get more done in a day.

The next question comes from Waqar Syed of Goldman Sachs.

Just some housekeeping. What do you think the average fleets -- active fleet would -- count would be for Q1 and then Q2?

It looks to be -- it's going to be roughly 20 fleets in this quarter, and we'll end the quarter with 21, and we'll go to 22 but only very late in the quarter so probably a little over 21 in next quarter.

Okay. And then for deliveries in the third and fourth quarter, they are -- you expect them to be at quarter-end? Or sometimes in the middle of the quarter? Or how should we be modeling those?

Well, the Q3 fleet looks like it's going to be right at quarter-end. Hopefully, it'll be -- should be in the quarter. And Q4 fleet sort of up in the air. That could be more at the middle of the quarter than to the end of the quarter, but it's far enough out, hard to say with any precision.

Okay. And for your -- the CapEx numbers that you provided for the year, could you provide a little more breakdown. You mentioned $55 million for maintenance and then $20 million from last year? Could you further provide breakdown in terms of between upgrades and newbuilds CapEx?

I'll let Michael speak to that.

Yes. Waqar, I mean, when we look forward to this. I mean, I think, on the ancillary service line, probably about between $20 million and $25 million is associated with that. We're also looking at about $13 million for our final refurbed fleet. And really, the balance -- then the balance of the CapEx being runs in these really newbuild, which includes pump-down equipment and the upgrades to high pressure.

Okay. So the returns on these ancillary services investments, are they kind of in line with what your returns are going to be on the pumping side?

They are, Waqar, they are. This is a double combination on -- in their oil, on these rig billing they have. They'll have stronger returns on their own rate, but they also will have the ancillary if they need to accelerating and making our other fleets more efficient, so yes, we expect it -- we're expecting great returns on those assets.

And that's the bigger reason, Waqar, why we're doing them. Yes, we'll have good returns on that equipment, but it's a relatively small business compared to the broader complex. The motivation to doing this, simply a hunt for efficiency, simply to figure out how to get more done in a day.

Okay, great. And then just one last question, do you expect to generate free cash flow this year?

Absolutely.

And could you quantify within a range, what do you think that could be?

I mean, I'm sure you've got a model as we've got a model of our financial performance for the year. Certainly, our expectations are to generate cash flow from operations above -- well above our planned CapEx, but I -- yes, probably not prudent to elaborate any more than that. But you've heard us from the past, or you've seen us in the past, we are return-on-capital and return-of-capital guys. So this year, maybe, mostly building and strengthening the balance sheet, but I think next year, you will hear us talk about return of capital.

The next question comes from George O'Leary of TPH & Co.

Just given, you -- all the focus on driving efficiencies for your customer and technology, maybe, as you pushed further into the Texas market, could you tick through some of the areas where your customers are acutely focused on improving their completions operations or various issues in the field? What maybe those issues are? And then how you guys are thinking about helping those customers?

You bet. I mean, look, as customers are moving. You see a significant increase in the number of pad drilling in operations in Texas. That means the ability to move faster is there. But water supply, wire line, maintenance of the well heads and other ancillary services, flow back manifolds to a perfect frac designs, you're often pushing the edge of what you can take. You've got to be able to deal with screen outs. Hopefully close it back and not have to clean out the well heads. But, the operators down there are savvy, very savvy. They've just been geologically and land acquisition focused, and then they've been delineating their acreage focus and now the focus for us and our customers is increasingly shifting to just smoother and more efficient operations. I don't know if my colleague, Ron, wants to elaborate on anything there as well or if I've hit the major points?

No, I think maybe the only other thing we're going be highly focused on there, of course, is this transition to regional sand. We anticipate that's going to play a significant role in completions in the Permian going forward. So we've been laser-focused on making sure that's something we can execute well on in partnership with our customers. So to that end, we've been out making sure that we have the appropriate supply chain in place there, geographical distribution of sand supply. So we want -- we worked hard to make sure we have access to regional sand in the north, in the south, in the east and in the west with access to a wide number of highways. Of course, you know we have proppants out there with the goal of giving us access to the widest possible trucking fleet. Any flatbed trailer, of course, can move sand for us. So we've been working hard on that to make sure that we're able to deliver that additional component as a partner with our customers as well.

And of course, it's always a balance sheet. You want throughput and efficiency. That's just straight up. But the other thing is to find the best way. The ultimate goal of our operations, and throughput in efficiency are large contributors in that goal, is to lower our customers' cost of producing a barrel of oil. So our engineering team has built, and very large database, now over 50,000 wells and very rigorous analysis to try to figure out how do you best do that. That's frac design, proppant volumes, mass sizes, completion types, fluid systems, rates. I could go on and on, but one of the targets of us right now is regional sand. We've been pumping regional sand since just past the middle of last year, and we are now doing an analysis of the impacts of that. Clearly, has this significantly positive cost impact. We've just got to confirm that productivity impacts are either non-observable, or they're so modest they're well more than offset -- on the -- than the -- they're more than offset by this decrease in well cost. That's an ongoing effort. And of course, that will be different, different parts of the basin and different horizons and all that. But yes, we're pretty bullish on the outlook for regional sand playing a large role there.

Great, that's very helpful color. And then kind of in the same vein, you guys have been super successful in kind of your legacy markets, most notably, the DJ Basin from efficiency standpoint and have put down some big frac stage per day numbers. As you progress into Texas, and I realize it's just relatively early days at this point but continue to grow in that market, maybe could you bifurcate the -- and I realize some of it has to do with the geology and the per stage volumes of sand that you talked about earlier. But could you bifurcate between how many stages per day you're able to get done in, say, the DJ for example, versus the Permian? And then can you close that gap and push the Permian closer to the DJ over time? Maybe just some color there would be helpful.

Yes, I think in pump hours or number of frac done in the -- frac activity done in a day in the Permian, even sand volumes, will they get to the DJ volumes? They will. On stages, will they get there? And there -- again, there's not really a DJ number. It -- we have customers in the DJ with -- probably at factor of 10 difference in sand volumes on Stage A versus Stage F. So it's -- we don't really -- I mean, we monitor stages, but stages aren't really the right unit to look at when you're talking about, can you have the same efficiency? I think pump hours in a day or pounds of sand placed in a day are better overall metrics for the efficiency or throughput in a day. Customers in the DJ or Bakken or whatever, we change stage sizes. Sometimes, we're doubling them. Sometimes, we're shrinking them. All those impact stages, but they don't really impact efficiency and throughput in a day.

Okay. That's helpful. And maybe, if I could just sneak one more. And I guess then, would you say the rate of change we can expect in hours pumped or efficiencies, kind of however you want to think about it, in Texas is maybe the rate of change should be higher there versus the legacy markets you guys have been more active in historically?

Yes, I think that's a reasonable assumption earlier on. So yes, a steeper ray of rise and improving efficiency. I agree with your statement, George.

The next question comes from John Daniel of Simmons & Company.

Chris, in your prepared remarks, you talked about seeing excess demand right now. Do you see customers expanding their frac fleet counts? Or are they simply upgrading their frac providers?

You see both, you see both. I think there is still a significant amount of expanding frac fleet count on customers. Even an admitted a frac fleet count for a while, their drilling efficiencies are coming up, their frac intensity might be migrating up. So in general, across the basin, there is an increase on the demand per frac fleets today right now. For sure, that number is going up. But certainly, for us, again, we're not really betting just on the macro (inaudible). To your other point, it's to find the right partners that will benefit the most and partnered the best in efficiency and throughput. So yes, some of them -- certainly, our fleets go into work but not only just because there's new capacity needed. They're upgrading operations.

Right. And I mean, you guys have done a great job from service quality, and you alluded to your -- the foundation the company being built on technology. But let's assume that the top true players, such as yourself, continue to successfully execute and take market share, this would seemingly suggest that your lower quality competitors will be forced to find a home for their fleets and may have to do so at lower pricing. And should that play out, you talked about having dedicated fleets, can you just elaborate on what you're doing to lock in utilization and pricing? And how you will approach that day when your competition, the lower quality folks, if you will, are forced to cut price to try to win back work? Just sort of a big-picture question for you.

Yes, you bet. So that's happening today. Even in great markets, that's always happening. There are different qualities of players, and they deliver a different service, and they get paid differently for that. It's more valuable to get the same frac stages done in the day if you get 7 done than if you get 4 done. It's your ancillary services for our customers, all their daily fees on a thing, they shrink. So do premium providers get paid a little bit of a premium price? Yes, because we deliver premium value. I think our customers win by that, too. Ultimately, it lowers their cost, even though we may get paid a slightly higher per-stage value than a lower quality player. So that dynamics of lower quality fleets getting displaced from customers they're in and having to find another home, that's been going on for -- I'm going to say the last 12 months, but that's been going on for the last 20 years. So there's always that in the backdrop. Now if you get to a situation where there's -- the frac demand for frac work has declined or a new capacity rolls on, exceeds current demand for frac fleet, and you have a different market dynamics, you still have a pricing differential between the operators. You still have a preference to have a preferred higher quality player on location. But now you're starting to see pricing pressure going the other way than it's going today.

Right. Okay. Final one, and more of a housekeeping for Michael. In your prepared remarks, you talked about an expectation for high single-digit growth in EBITDA for Q1, which, again, based on my dumb math would suggest EBITDA just shy of $100 million. Is that fair?

That's how the math works. Yes, John.

Okay. But when you -- with that math, is that including an estimate for equity-based comp? And if not, can you give us sort of a range for what equity-based comp might be for the year?

It -- we probably could take that offline. We're still working on the (inaudible) equity-based comp at the moment.

The next question comes from Scott Gruber of Citigroup.

This is Bart Rekucki just stepping in for Scott. So can you guys just discuss your longer-term growth plan? I know you previously discussed entering Midcon or Appalachia. Is that still in the picture? Has that changed at all?

No, I'd say no change in our broader plans. And again, just like with the fleet deployment, it's a little bit more focused on customers than basins per se, but there will come a time where we decide to enter an additional basin. For sure, we'd be driven by a customer or 2 customers that we like, and we love the opportunity, and we want to help them out. But that's a bigger decision. We've got to build a base. We want to quickly get scale there and all that. So surely, Bart, we will do that. I -- it's not in the near future. It's not right through the windshield right now. Our market share in the Permian and Eagle Ford is still small. The Permian is a huge basin. We're a new entrant in the Eagle Ford. So for the -- for a while, we'll be focused on the basins we're in. But we are constantly building request and dialoguing with customers. And that day for when we had the next basin, it'll come. I can't tell you when, but it's not in the next 12 months.

Got you. And for the fleets that are going to come in 3Q and 4Q, can you just talk a little bit about visibility on -- do you guys have like a handshake agreement with customers for -- that used to go to work? Or is it just more in the discussion at this point? And do you think that would be for existing customers or new customers?

It's -- we are in discussion. I think we know where the fleets are going, but there are other people we're talking to, so we're -- we haven't formally clarified that, but I'd say we have a pretty high probability of where those last 2 fleets will go. I think both of them will likely go to existing customers.

The next question comes from Vaibhav Vaishnav of Cowen.

Chris, I believe you mentioned return of capital. I was hoping -- if you can, provide some color around how you think about that.

Yes. In the past, we've done it through private companies. And there, it's simply special dividend. When we generate excess cash during good times, we'd mail up per share dividend to all the holders. In a public marketplace, we have an additional option, which is share buybacks. And so ultimately, it will depend when we're prepared to do that, what's the valuation of the stock. If we think this stock is significantly undervalued, we may deem that the greatest way to accrete value to our shareholders is through a buyback. If it's closely reasonably valued, it's more likely we would do special dividends. But our goal, I mean, what motivates all of us and as throughout our careers is to increase the value of a share stock. We've been large owners in every business we've been in as we are here, and to me, that's the purpose of a business, grow the value of a share.

So it sounds like you guys, based on your internal model, I think you have free cash flow -- significant free cash flow at the CFO much more than CapEx, as you stated in your comments. Going forward, is the best use you're going to be more new bills? Or is it going to be more of a function of buybacks?

It absolutely depends on the circumstances. There will be circumstances where we may have no new bills in a year. And all of the free cash flow, if it's going to be -- I mean, maybe distributed to shareholders. There'll be years like this year, where we're establishing a presence and trying to change the game a bit in the Permian, where the majority of our free cash flow is going to. CapEx, an expansion of our business. So that's a -- again, ongoing dialogue with our customers. So it's granular. It's macro, where we think we are in the cycle, how do we feel about our robustness for the inevitable downturn. It's not in the windshield right now, but they always come, and it will come again. So for us, the proportion between those things will never be fixed. It will swing significantly between growth capital and returning cash to shareholders.

That's helpful. If I think about March, you said like March looks -- sounds like it's more normal weather, normal operations. If -- can we think of March exit rate around that $22 million that we saw back in third quarter '17 per fleet EBITDA?

I prefer not to do that. Of course, we're not even halfway through March and all that, but things are going well. We feel good about where our business is right now and the outlook, but we'll have to provide a more specific guidance than that. Appreciate your sentiment. I know what you're looking for, but I'm going to hold back.

No, I completely understand. And thinking about fleet, fleet number 23, 24, sounds like pricing is still improving. I'm not sure if the pricing is already locked for those 2 fleets, but sounds like that should be at least at this level or higher. Is that a fair way of thinking about it?

Yes, market -- if the market is strong, then yes. If there's a pricing change going on, it's still a slow drift upwards. Not the rapid change as we saw last year, but there's still more demand than supply. And yes, so pricing is slowly -- continue to drift upwards.

And last one for me. Sorry, I apologize if I missed this. Did you guys mention like, for 2 newbuilds, what's the dollar per horsepower cost?

It's -- as we always say, dollar/horsepower for newbuilds is approximately about $1,000 of horsepower, very slightly above that with the new one. Ron can maybe talk to a little bit to the new design of the fleets that's coming out at the moment.

Yes. I think we've always said that we've never been a company that's focused on lowering the upfront cost of purchasing a piece of equipment. We've always been about total cost of ownership. So in this particular case, we are looking at a next-generation horsepower unit that's coming out. We'll call it Quiet Thunder. So it will be a next generation of our Quiet Fleet but with a new generation pump on the end of it. So we've seen pretty significant progression in the maintenance or rebuild cycle for pumps and -- or engines and transmissions. We've got that number pushed up to 10,000 pump hours kind of thing. And really, the piece that wasn't aligned with that was the power end on the pump. This next-generation pump will get us to a point where all of those major components are now aligned. So engine transmission and power end will have a rebuild cycle, all in and around 10,000 pump hours, call it, 15,000, 16,000 engine hours or something like that. So we're extremely excited about that and look forward to getting these next couple of fleets out there to see how they perform.

Yes, and just to follow one last thing on Ron's comment. To us, it's always -- which is why we added this metric, it's always about return on capital employed. If I can buy something for half the price, but I'm only going to make 1/3 as much money with it, that's not cheap, that's expensive. So we're always looking at how to -- to accrete value in an organic growth company, you've got to reinvest capital on a high rate of return. And so that's, again, total cost of ownership, total return on an investment. That's how we look at every dollar we spend.

Thanks so much. Great questions from everybody, and I believe our Q&A queue is empty.

Yes, and so I thank everyone very much for joining us today, for your interest in Liberty and for the thoughtful questions and dialogue.

We're excited about where we stand in 2018. We're very excited about the partnerships we have with our customers, very proud about the success of their operations.

And frankly, the 3 of us here are very proud to be on the Team Liberty that's -- that loves what we do, and we look forward to talking to you again in the future.

Everyone, have a great day.

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.