Liberty Energy Inc (LBRT) 2018 Q3 法說會逐字稿

Good morning, and welcome to the Liberty Oilfield Services Third Quarter 2018 Earnings Conference Call. (Operator Instructions) Please note today's 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 pretax 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, adjusted EBITDA and the calculation of pretax 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, and thank you for joining us. We're pleased to discuss with you today our third quarter 2018 results. In partnership with our customers, the Liberty team continues to drive high efficiency operations, which are a win for Liberty and a win for our customers.

Strong cash generation in the third quarter enabled us to execute on returning $60 million of cash to shareholders in the form of a regular quarterly dividend and repurchasing 2.4% of our total outstanding shares while reducing our net debt to $20 million.

In the dynamic market conditions that we are in, customer relationships are paramount. Fortunately, Liberty's whole business is built around customer partnerships. As our customers' plans change or market conditions change, we work with our customers to adapt. As we mentioned in our last call, Q3 began with customer scheduling challenges, and they continued throughout the quarter in various fashions.

Despite these challenges, our third quarter results were strong, with revenue of $559 million and net income was $66 million or $0.49 per fully diluted share. Adjusted EBITDA for the quarter was $117 million or $21.2 million per average active frac fleet on an annualized basis. Working in concert with customers, Liberty continues to drive innovation and operational efficiency across the entire fleet. This performance translates to strong demand for Liberty's high efficiency fleets that deliver differential frac services.

Premium service quality, coupled with basin and customer diversity, positions the company to believe that it will continue to generate strong returns on capital employed regardless of how the market unfolds in the next few quarters. Liberty was built for long-term success as illustrated by the 12 months' pretax return on capital employed of 43%. Liberty's geographically diversified operations continue to be in high demand, and our long-term partnership strategy allows us to work closely with customers to make the most efficient use of our assets and expertise to lower our customers' cost of production.

During the third quarter, additional local sand volumes continued to come online, driving down well costs for our customers. We have ramped up to pumping over 70% local sand in the Permian Basin currently.

The frac pricing environment has been weakening modestly in the second half of 2018 as pressure pumping supply that was built for expected Permian completions growth outstripped the flattening completions growth curve. In a normal year, fourth quarter revenue typically declines mid-single digits sequentially due to the holiday season and budget management by producers. Current indications for the fourth quarter suggest that this is not an unreasonable expectation for this year. Global oil markets remain constructive. The rapid pace of inventory draw has slowed significantly but days used in storage reflect relatively normal inventory levels.

Markets are focused on the loss of Iranian oil due to the return of sanctions and continued decline of Venezuela's oil output. These significant output declines are being roughly offset by increases in oil production from Saudi, Russia and the United States. Today's $65 to $70 oil prices provide strong well returns for our customers.

With a supportive macro commodity environment and the projected increase in takeaway capacity coming online in major basins, we would expect the supply-demand balance for frac services to tighten and pricing to strengthen in the second half of 2019. Demand for dedicated efficient fleets looks to be strong in 2019. For the entire third quarter, we ran 22 active frac fleets. We anticipate deployment of our 23rd and 24th fleets in the first half of 2019. These fleets are currently under construction.

As an example of one of our recent efficiency-focused technical efforts, Liberty has focused on the dynamic sand logistics environment. Despite increasing use of in-basin sand, more than 60% of the sand Liberty pumps next year is expected to move via rail. We will make more than 300,000 truck trips hauling sand. We are developing the next generation of our software to streamline the management of this process. The end product will provide complete visibility of our profit supply chain, including rail, transload and last mile in one platform, all integrated with our ERP system.

Geo-fencing, route monitoring, eBOLs and integration with e-log technology will provide the ability for optimized last-mile dispatching across each basin with a strong focus on safety. Ultimately, these improvements increase our efficiency and benefit our customers. And this is just one example of our relentless pursuit of improvement.

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

Good morning. We are pleased with our third quarter 2018 results. The Liberty team worked tirelessly to provide exceptional execution for our clients and deliver strong financial results in the face of customer scheduling choppiness. For third quarter 2018, revenue decreased 11% to $559 million from our record $628 million in the second quarter of 2018.

Net income totaled $66 million in the third quarter compared to net income of $95 million in the second quarter. Third quarter adjusted EBITDA decreased 21% to $117 million from $149 million in the second quarter. Annualized adjusted EBITDA per fleet decreased to $21.2 million in the third quarter compared to the $28 million in the second quarter.

Although annualized adjusted EBITDA per fleet was down from our record setting second quarter, we are very pleased with the third quarter earnings numbers, especially in the face of the scheduling issues the team had to deal with. As we have said, we are a returns-focused company and at the end of the day, sustaining cash flows from investment are what drives returns.

Sustaining cash flow per fleet is a metric we use to measure through-cycle fleet profitability. It is an important metric we use as an input to deciding future capital commitments. We define sustaining cash flow per fleet as the expected annualized adjusted EBITDA per fleet less our expected annual maintenance capital per fleet. Through the third quarter and year-to-date annualized adjusted EBITDA per fleet was $23.2 million. As previously announced, our expected annual maintenance capital for this year is approximately $2.5 million per fleet.

General and administrative expenses, excluding $2.2 million of fleet activation costs, totaled $22.5 million for the quarter or 4% of our revenues. Third quarter G&A includes stock-based compensation expense of $1.6 million. Interest expense and associated fees totaled $3.6 million for the quarter.

Third quarter income tax expense totaled $12 million compared to $16 million in the second quarter. Liberty was not subject to income tax prior to its initial public offering. For the remainder of 2018, we expect our reported income tax expense to be approximately 16% of pretax net income. For the fully diluted earnings per share calculations, our effective tax rate would be 24%.

We ended the quarter with a cash balance of $87 million and total net debt of $107 million. At quarter end, we had no borrowings drawn under our ABL credit facility and total liquidity, including availability under the credit facility, was $337 million.

As we discussed previously, in order to seek the best long-term returns for our shareholders, we will follow a prudent strategy of maintaining a strong balance sheet, investing in compelling growth opportunities and returning capital to shareholders when appropriate. In the third quarter, we paid our first quarterly dividend of $0.05 per share and announced authorization for a $100 million share repurchase program. During the third quarter, we repurchased 2.8 million shares, reducing our total outstanding share count by 2.4%. As of September 30, 2018, the total remaining authorization is $46 million.

Additionally, our Board of Directors announced on October 23 a quarterly cash dividend on our common stock of $0.05 per share to be paid on December 20, 2018, to holders of record as of December 6, 2018.

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

Liberty's strong financial results, favorable outlook and strong balance sheet support our balanced strategy of growth and returning capital to our stockholders. Liberty is committed to creating long-term stockholder value via compounding shareholder value by reinvesting cash flow at high rates of return and returning cash to shareholders as appropriate.

We are excited by the growth opportunities in front of us and the positive long-term outlook for the shale revolution and the benefits that this brings to our industry and the country as a whole. Thank you for joining us, and we'll open up for Q&A now.

(Operator Instructions) Today's first question will be from John Daniel with Simmons Energy.

Chris, you and your team are regarded as one of the more forward-thinking players as it relates to technology and innovation within the frac market. So with growing signs that E&P customers are exploring electric fleet technology, can you say what your thoughts are on this and when and if we'll see you guys pursue a similar strategy?

Yes, John, I believe we spoke on it last time. And funny as it may sound, we've been looking at electric frac fleets since we started this company. We're engineers by training and so we're interested in that. But to date, the 2 biggest advantages of electric frac fleets is they're dramatically quieter than a standard frac fleet, and they burn natural gas instead of diesel so they reduce fuel costs. But at Liberty, several years ago, we developed these quiet frac fleets. They're as quiet as an electric frac fleet. And about 40% of our capacity is dual-fuel, which means that with the newest generation of that, we can get almost 90% of the energy to run those engines from natural gas, 11% supplemented diesel. And then when you have intermittency, you have the robustness to keep running operations. So we continue to look at electric frac fleets. But right now, the math of the cost of them and the requirement that you have robust, stable delivered natural gas or operations shutdown, they just don't look as attractive as the technology we have today. But we will continue to look at them. And as soon as they make sense or they're close to making sense, you'll see Liberty in that space.

Okay. One follow-up and I'll let others hit on the modeling questions. But with the dual-fuel fleets, Chris, do you see any demonstrable improvement in the fleet profitability for those, that 40% your fleets versus the others, or is it still about the same?

Ultimately, there -- most of the advantages of dual-fuel goes to our customers, right, because it saves fuel costs on their operations. And again, compared to other efficiencies and things we do, these -- they're meaningful but they're not dramatic. But yes, I mean, we probably have a larger percent dual-fuel capacity than anyone out there so does it open up or favorably position us to a few more opportunities having so much dual-fuel capacity? It probably does. Is it needle-moving or noticeably different profitability? Probably not.

Next question will be from Sean Meakim with JPMorgan.

So Chris, your point on seasonality in the fourth quarter is well taken, but should we think of that mid-single-digit revenue decline as a midpoint of some type of range, I'd be inclined to think that's more towards the upper end of the range. And then also, given that revenue is less relevant today as we're just having this mix towards local sand, I think the more important metric will be how do we think about the range of EBITDA per fleet in 4Q and maybe your confidence level in 4Q being a type of bottom for that measure.

All right, Sean. You're right, that the seasonality, which is both budget management and it's holidays and people shutting down so it's always a bit of a challenge. I think we're another year older, another year deeper relationships, so I think we've got probably better views into our customer plans further out this year. So we've been able to plan and fill gaps where people have decided to stop early. We've been able to plan a little bit better on that, which that probably is offsetting a bit of a natural, a little bit of revenue decline from greater use of in-basin sand. But I think we feel pretty good going into Q4 and where we stand and how busy all of our fleets will be. Our guess is that the decline in EBITDA per fleet in the fourth quarter is probably around $3 million per fleet decline from Q3, maybe a little less than that. And our outlook certainly makes it look like Q4 is the bottom. We're in multiple discussions right now about Q1 in next year, and we are back in a situation we're typically in, which is more demand for our fleets than we have fleets starting in January. So we're trying to stay loyal, protect the long-term customers first. And I think we will cover all the people we're deeply in conversations with. But the start of next year looks pretty strong for Liberty.

That's very helpful. And then I was thinking about the fleets 23 and 24, and those deployments now going in the first half of '19. Any changes in terms of customer commitments or where they're expected to operate, and I'm thinking, is there appetite from your end to deliver incremental newbuilds next year beyond those 2? Or given the expected ramp in '19, would that maybe something, a decision point, for mid-next year for 2020 deliveries?

We monitor that closely. As everyone knows, right, the market in the Permian, there's tons of frac work going on but the market softened a little bit there. There's a number of idle fleets. So we have customer interest to bring new fleets into that market today, but the -- sort of the pricing today is weaker there. They're not quite as impacted by all that in the Rockies, so we have more than 2 hands up for those 2 fleets that we'll finish in the first half of the year. It is very possible that both of those go into the Rockies. We expect Permian, or turn up in activity, to be more second half-weighted, although the dialogues about people's plans, they are going to be late this year, they're going to be early next year. And look, customers, we've got that we're close with, they've got increasing capacity. It's certainly possible that we would add another fleet or 2 next year. Remember, those fleets 23 and 24, they were budgeted to add this year, but we slowed them down in expectation of the market. We were able to juggle some things and cover the customers we had committed to for this year.

Next question will be from James West with Evercore ISI.

Chris, with fleets 23 and 24, sounds like demand for those is strong. I know you are trying to time the market as appropriately as you can at this point. It doesn't sound like they're dedicated to somebody yet but should we expect either announcement or as those come to the market as the construction is finished, that they would go to work basically immediately?

I think for the most part, they are dedicated to someone already. We're just looking at the scope of other customers' plans next year and whether -- where that fits into existing capacity. So there could be some movement, but I think it's pretty clear they will be fully utilized when they deploy. The issue, James, is more we probably have got 5-plus dialogues for those 2 fleets and really, those 2 will go to the first 2 that start up under good term, we think we know who those are. And we'll continue dialogues with others. That's why I said it's likely we add another fleet or 2 next year. We won't add a lot of fleets next year. But there is going to be meaningful ramp-up in West Texas next year. And in the Eagle Ford, we have a lot of demand and inquiries for what we do. So we're continuing the usual dialogues but with existing customers slowing down a little bit, it just pushes off a little bit the urgency to need to add capacity.

Okay, fair enough. And then a follow-up for me on your sand suppliers. I believe you had some contracted sand but it looked like in-basin sand ramped up nicely throughout the quarter, helped you on the cost side. Can you just remind us what your contract status looks like for your sand supply?

This is Ron. From a contract standpoint, we've always had a philosophy where we aim to have some amount of our capacity roll off every year. That was probably around 30% so maybe kind of a rolling 3-year sort of average on contract length. And so what that's meant is that we -- as we ramped up in-basin capacity this year, we've just allowed some of our existing Northern White contracts to roll off as they reach their termination point. So we've had a -- we're in a good situation where we've been able to balance what we've been taking on from an in-basin standpoint with the end of Northern White capacity we'd used in the past.

Next question will be from Jud Bailey with Wells Fargo.

Question, Chris, on your -- the EBITDA per fleet. I think you said around $3 million or so. Is that going to be more price or is that expectation of slowdown in pumping efficiency for the holidays? Or could you help us think about what the mix is that is pushing that down? Is it more price or is it more expectation of slower efficiency because of the holidays?

It's mostly customer schedule efficiency. The big drop from Q2 and Q3 was just there were more days that fleets weren't pumping. That is the dominant driver of the decline from Q3 to Q2, and it will be the main driver in the decline from Q3 to Q4. It's just when you have people -- you got -- think you've got a customer that's sort of ramped up and finished up their spending for this year, I'd like to think part of that's due to Liberty's efficiencies, but they ramp up their -- I mean, they end their program in end of October, right? You've got 2 months or so you've got to fill. We've got people interested in that, but it's not the same crew working on the same wells and the same things. So, there's a loss of efficiency, and we've got to start when the next customer wants to go, so there's usually a gap between those 2. There's a ramp-up efficiency getting going and starting in a new place. So it's mostly less days fracking and a little bit less efficient operations when you first ramp up and start a fleet with a different set of -- a different customer and a different location.

Okay. That's good color. And my follow-up is on pricing. As you look into early '19, how are you in terms of pricing reopeners for your contracts? I know that's usually a pretty fluid situation between you and your customers. Do you feel like you -- have you already kind of created some price concessions, I guess? Or are those to come in early '19? Or how do we think about your portfolio of contracts and kind of where pricing is and where you think it could be in the following months?

Yes, Jud, so I agree with your comment, for us, it's very fluid. We've always had very much a partnership mentality. We agree typically with a customer that we're going to do X amount of work for them for the year so there's a plan for it. And then pricing changes as markets change. Sand prices go down and fantastic. That's good for us, that's good for our customers. We adjust it immediately as it happens. If there's excess frac fleets out there and sort of service pricing in a softer market is softening meaningfully, we adjust our pricing as that happens. Up and down. So, no, we -- as the market has softened, we've moved pricing down together with our customers. The bigger driver of lower frac stages cost for our customers had been decrease in material costs. And just as the relationships season, we get things done more quickly, which lowers that first-stage cost for customers as well. So we don't expect any -- there's no big cliff change that's going to happen on January 1 for us in pricing at all. In fact, again, it's -- there's less demand for frac capacity in Q4 than in Q1 so if there was any movement in price it might be the opposite direction. But I would expect pricing actually relatively flat from what it will be in Q4 and what it will be in Q1. But I think we'll have a better, more efficient higher throughput schedule in Q1. But pricing for us is dynamic. It is not based on -- we may have 2 fleets in our entire thing that are like firmly contracted for a full year at a set price. Almost all the others, they're agreed upfront, but as market conditions change either way, we talk and we adjust as it happens.

Next question will be from George O'Leary with Tudor, Pickering, Holt & Co.

I noticed the commentary in the press release and discussed today as well, just that pricing in the Southern basin, you've been a little bit softer, which makes sense, given what we've seen with activity trends. But I wondered if you could speak specifically to the Permian versus the Eagle Ford within your Southern operations and what you're seeing from a pricing and utilization dynamics perspective in both of those markets.

Yes, look, the Permian if you go back 6 or 9 months ago, it was sort of the opposite end of the spectrum, right? Demand was growing very fast. People were adding new capacity. But all the fleets that weren't able to find work in the Marcellus or the Bakken or the Eagle Ford, or the Haynesville, they couldn't find work there. Where did they go? They went to the Permian because it was the strongest market, the greatest mismatch, in this case, more demand than there actually was supply. So now I would say that greatest gap for demand today versus in-basin fleets, that's greatest in the Permian right now. I don't think the Permian market's going to be terrible. I think this is a, whatever, 2-, 3-quarter where things are softer. If you've got a great high efficiency fleet, you're going to keep working. But most of the fleets that are not idle -- that are idle, I mean, they are idle for reasons. And so geographically, if you're in the Permian, is it easier to drive over to the Eagle Ford than the Marcellus, I mean, sure. So I would say -- and where that impacts, I should be clear, where this impacts the most is spot, right? So you're trying to fill 2 months or you're trying to do any work you can on the spot, the spot market pricing swings much higher and it swings much lower. So I should qualify all of this with, we don't play in the spot market, so we hear all this stuff, and of course, the same people looking for work in the spot, they throw in bids for dedicated fleets, too. So there's market communication but we're not a spot player. But yes, is it a little softer in the Eagle Ford than maybe basins further away from the Permian? Probably. But again, markets are dynamic and fleets move. And I don't think they have a long-term structural difference in where profitability is, but we've got a short-term softness in the Permian. We're not in the Marcellus-Utica, but it sounds like they've got a similar issue there.

Right, that's very helpful color, Chris. And then I apologize if I missed this, I was jumping over from another call this morning. But spreads 23 and 24 are those already delivered today and it's just the activity is kicking in, in 2019? Or did you guys actually kick the delivery of those a little bit down the road? Just curious from a mechanical perspective how that's actually played out.

They're not delivered today. I think we mentioned before, to get exactly the configuration we wanted, we had some delays on that stuff, so it was just delayed originally by getting the right components. So no, both fleets are currently under construction. Now could we have rushed them and swapped out some less desirable components that we'll swap at a later date? We could have. But given the market conditions, it didn't -- it wasn't necessary to do that. They're both under construction. They'll both be done in the first half of next year and they'll both be deployed in the first half of next year. So we did mention earlier, yes, there's significant demand for those 2 fleets, so we're quite confident that they will both go to dedicated work as soon as they're done and as soon as they arrive.

Great, I'll sneak in one more if I could. Those fleets 23 and 24, you talked about maybe having 5 guys who are interested for those 2 fleets. Does it -- do you feel more likely to put those with incumbent customers or are you looking to new dancing partners?

There's always a bias to incumbents, there's always a bias. If we have incumbent partners and they have a strong need and we have a meeting of the minds there, that's always where they go first. And both of those fleets will probably be with incumbent players. But part of it is maybe someone's got a fleet today and they need a 1.5 fleet, it could be to fill a gap there and we fill the other half with someone else. But right now, what we're really trying to do is getting exactly customer plans for next year and exactly how much capacity that takes. But our customer profile and customer makeup next year will not be meaningfully different than it is this year. It's not meaningfully different today than it was 6 months or so ago. So [there will be] a more gradual addition of new customers but that's not the main driver of growth.

Next question will be from Connor Lynagh with Morgan Stanley.

I'm wondering if we could get your take on the Proposition 112 initiative and just sort of your thoughts where things are heading, what your response would be if things go the wrong way for the industry. Just generally your take on where that stands.

Well, I'll start right off with saying that we are not in favor of Proposition 112. And together, if you want -- the industry in Colorado, which is a pretty tight-knit community, has come together in many, many ways to educate the State of Colorado, the leaders in Colorado and the folks on what this issue is, what it means, what we do as an industry. I think that is one of the positives that will come out of all this is we've spent a ton of effort reaching out in community -- communicating with the broader community, something we haven't done enough in the past. We had a rally 2 weeks ago with the mayors across Colorado and, believe it or not, the mayors of the 10 biggest cities in Colorado, all 10, have come out against Proposition 112 and either spoke at the rally or sent a written statement passionately on why they oppose this measure. So we put great efforts in educating the population of Colorado through multiple avenues. And by all the indications or data we have, those efforts have been effective and they continue to be effective. So we feel pretty good about where we stand right now. But the votes are counted in 6 days.

That's fair. And I guess, how should we think about if things were to go the wrong way, I mean, is there a backlog of work that you guys are going to have regardless? Do you have some time -- like what would be the impact on your operations just broadly?

Well, in the short term, nothing. But yes, there is -- there was actually a rather large permit backlog in Colorado. I think it's estimated anywhere from 1.5 to 5 years. So yes, the DJ Basin doesn't turn on a dime. If it did pass for sure it's a negative, for sure there's a scramble, there's a change, of course, it will be challenged in the courts, you can't just take everyone's property via a democratic vote. There will be a battle about that. But activity level will not change right after election day. But if it went the wrong way, if things went poorly, there will be a decline in activity but probably pretty gradually. And for Liberty, I mean we will fight to the end. But for Liberty, all our assets are on wheels and I think we would have plenty of time to orderly deploy them elsewhere.

Next question will be from Scott Gruber with Citi.

What level of CapEx is left on fleets 23 and 24 that will hit in '19? Is there much left?

Yes, Scott, we've obviously taken the deposits on the equipment and some of the ancillary equipment is being delivered. So about $60 million of the CapEx will roll over into 2019 that was planned for 2018.

Was that $16 million or...

6-0, $60 million.

Okay, got you. And was fleet 23 staffed during the quarter? Was that the startup cost that hit in the quarter?

That's hiring engineers. That's -- we do hire engineers way in advance. We do hire some crew leaders or move crew leaders into different positions. So yes, we are constantly refreshing people for that. And so yes, yes, yes, a good chunk of the hiring for fleet 23 is done.

So you also just have to hire for fleet 24 early next year?

That's correct. That's correct.

Got you. And then I heard your response on the pricing side. But if we just put pricing to the side, Chris, as we think about the activity trends, as they start to improve early next year, where do you think EBITDA per fleet could go in 1Q and 2Q if you want to opine upon it? But based upon your expectation around activity improvement and the mix of that activity, where -- what's a good starting point for us to think about EBITDA per fleet in 1Q and 2Q?

As you've probably seen in the past, we're not big crystal ballers. So it's -- I would say outlook where we feel pretty good about Q1, you look today at Q4, now we talk about numbers here because we're 1/3 of the way through the quarter already. We pretty much know where everything is going to be every day. So that's an easier thing. So I don't know, I'll turn it over to Michael. But it's probably best to wait to see how it unfolds, but I think we'll have a -- we're -- we feel pretty good about 2019 and that's all 4 quarters.

And I agree. I mean, as we've commented in the press release, we think the pricing, kind of the market is going to tighten, the second half of the year pricing is going to improve. And obviously, we've said that we think Q1 -- we think Q4 will probably be the bottom so Q1 will be of order the same or above Q4. So I think, from me, you can extrapolate.

Got you. And one big factor that the investment community has been debating around when that pricing power pendulum swings back your ways is efficiency. You guys are obviously an efficiency leader in the industry. What are you guys seeing, if you put to the side just some of the slowdown around pipe constraints and seasonality, but on a year-over-year basis what type of efficiency improvements are you guys seeing and where do you think that goes in 2019?

It continues to rise. It's for a number of different reasons. Some is new technologies, things that we develop that just save minutes on location. A lot of it is just aligning and planning better with customers, there's third-party services, there's coordination. But no, I think certainly within the Liberty family, efficiency will be higher next year than it is this year. And we'll -- and we hope and I believe we'll say that's going to be true every year. That's a major factor in choosing the customers we align with. I think it's a major factor in customers choosing -- our customers choosing to align with Liberty. And yes, our goal is to keep getting done more every day with every fleet we've got. And so yes, I think that trend continues.

Do you think it slows down at all? Do you think you keep pace? And is it at a double-digit type pace of increase in '19, if you would ballpark it for us?

No, I would doubt that. Look, obviously, at the start of our business, you pick the low-hanging fruit first, right? We were the odd guys that showed up, we show accounting just for the minutes of every day. That didn't seem like that should have been rocket science. But I would say at the beginning, we had a lot of low-hanging fruit. And as you get further into it, there's still plenty of room for improvement but you tend to hit the big ones first. So yes, the weight of improvement of efficiency, yes, that probably slows with time.

Next question will be Blake Gendron with Wolfe Research

First, and I apology -- apologies if I had missed it earlier. But comments from key customers it seems in the DJ about CapEx potentially pulling back in 2019. It seems like you have really good visibility into the fourth quarter, but as you think about your business in 2019, is there some sort of our contingency plan? Are you not worried about the backlog of work for '19? It seems like we've heard some good things activity-wise out of the Powder and obviously, the Bakken is close by. How do you think about maybe the CapEx guidance from key customers?

Yes, from what we've heard so far, I think CapEx guidance in general for our key customers next year will be up. Their production is up and oil prices are up. So yes, I think we feel pretty good about growth in our business next year. Our customers in general feel pretty good about their growth. But that's not every customer, right? Some are going to shrink, some are going to grow, but when you average across it, I think in our customer universe, next year will be up.

Okay, great. And then for the Texas fleets, are you still in the process of aligning with customers? And are you switching customers to try to find that right fit? Or have you identified those customers that you plan on galvanizing a longer-term relationship with?

I would say mostly aligned right now with customers. We're talking to tons of new ones that have heard the Liberty story, we're showing things and talking so there's always a dialogue. Because look, we're newer in that basin, we're going to grow a lot in the Permian Basin. But the customer base, that uses Liberty capacity today in the Permian, yes, not meaningfully different than it was 6 months ago. Our key partners are our key partners. So yes, so less dynamic than it was 6 months ago, a lot less dynamic than it was 12 or 18 months ago.

Okay, great. And then last one for me. I guess, just refresh us if you can on how you think about potentially other basins, namely SCOOP, STACK? Or are you just focusing on rounding out your scale in existing basins?

Yes, I think where we are today and we look at into next year, the dialogues with existing customers and existing basins, there's well more demand than we will, from a CapEx perspective, be willing to supply next year. And we're still newer in the Eagle Ford and not that old in the Permian. So we will not expand to a new basin next year. I think another year to season and grow and build our technical expertise in those basins is reasonable. We are constantly reached out to by folks that are in the Haynesville or SCOOP, STACK or the Northeast. We got people come down from Canada and talk to us about the Montney. So I think with time, you will see us move to additional basins, but I would say slowly and methodically, and not next year.

Next question will be from Stephen Gengaro with Stifel.

Really just 2 quick ones. One, when you think about the move in EBITDA per fleet in both 3Q and as you think about 4Q, is there a way you can help us understand, and I think it's mainly utilization but, what the price/utilization impact is on the moves in that number?

Yes, Stephen. As we've said, I think, by far and away, the biggest driver is utilization. We've expected some customer choppiness in Q3 down from a very, very, very highly utilized, as you said, in Q2, everything lined up and went perfectly. So a good portion of that drop was natural because things don't go perfectly every quarter. And then in Q4 really, we do always get this sort of holiday slowdown, you have the holiday breaks and then comes budget management at the end of the year. So yes, by far and away, the dominant portion of that is utilization of scheduling. And as we've said in the southern region, there's been some price weakening, slight price weakening but across the board, it's in the low single digits.

And then as you think about your customer relationships, and you mentioned, it kind of moves in price as the market bounces around on the spot market, how tightly are sort of the price moves in -- under your arrangements? Are they relative to the spot? Are there certain moves that could be made on a quarterly basis? How should we think about that?

No, they're not relative to the spot. But obviously, sort of markets, it is -- there is a market for frac services, right? So what you'll find is we have long deep relationships with our customers. And I think you'll see over time, as frac prices go up, we will not sort of like push the boundaries at the top end of frac pricing and we never ever plumb the lows of the bottom end of the general market. So these markets move -- our pricing moves gently in conjunction with the market other than commodity prices. When sand or some major chemicals move, we pass that straight through to our clients and those savings go into helping their well economics.

Great. And then just one final quick one. When we think about maintenance CapEx per fleet per year, any changes to sort of that $2.5-ish million number?

No, I don't think so. I mean, as we've said, I think we said about a -- when we were on the road about 9 months ago, we would expect that to sort of like move up slowly over the next couple of years to around $3 million a fleet or just a bit above that, probably over the next 2 or 3 years. But that's a good number. But we have a very conservative capitalization policy, we only capitalize the maintenance on the major refurbs of our 3 major items on the fleet, everything else is expensed. So yes, I would not expect to see that change markedly.

At this time, this will conclude today's question-and-answer session. I'd like to turn the conference back over to Mr. Chris Wright for any closing remarks.

Yes, I just want to say thanks to everyone that listened in on the call and for the great and thoughtful questions we had today. And thanks to everyone in the Liberty family, the broader Liberty family, that makes us love our jobs and love being a part of the shale revolution. That's folks that work in Liberty hard every single day. We've got a wonderful passionate crew. We've got a fantastic crew of customers who've become our partners and fantastic suppliers as well. But all of those 3 entities together have to deal with the dynamic market and we do it together as humans and love what we do. Thank you for your time today.

The conference has now concluded. We want to thank you for attending today's presentation and at this time, you may now disconnect.