Nov 12, 2019
Operator
Welcome to the Q3 2019 Air Transport Services Group Incorporated Earnings Conference Call. My name is John, and I'll be your operator for today's call.
[Operator Instructions] Please note that this conference is being recorded.And now I will turn the call over to Joe Hete, Chief Executive Officer, ATSG. Mr.
Hete, you may begin.
Joe Hete
Thanks, John. Good morning, and welcome to our third quarter 2019 earnings conference call.
With me today is Quint Turner, our Chief Financial Officer. Rich Corrado, who became President of ATSG in September, is traveling overseas today.
We issued our earnings release yesterday after the market closed. It's on our website, atsginc.com.
We will file our Form 10-Q later this week.Once again, I'm pleased to report to you that this year, we are performing at levels consistent with the adjusted EBITDA guidance we have shared and exceeding our targets for adjusted EPS. Our business strategy of investing in and leasing midsize aircraft to major express and e-commerce customers and combining those assets with crew, maintenance and insurance services for those who require them continues to be a winning formula.
Our purchase of Omni a year ago has continued to provide accretive contributions to our operating results.For the third quarter, we again substantially outperformed our prior year results. Our revenues are up 79% to $366 million.
Our adjusted earnings are up 10% to $21 million or $0.31 per share, and our adjusted EBITDA rose 47% to $109 million. Our airline operations are expanding, and our leased aircraft fleet is growing.
Our air network services for the Department of Defense and Amazon are up versus the prior year, and we have 19 more aircraft in our operating fleet than a year ago. These results, plus significant increases in scheduled peak-season flying, support my confidence in our ability to achieve our $450 million target for full year adjusted EBITDA despite a few headwinds to our plan.Quint is ready to review our consolidated results, including the release we also issued yesterday about a new amendment to our senior credit facility.I'll close with comments on our outlook.
Quint?
Quint Turner
Thanks, Joe. Good morning to all of you on the call right now and to those who will listen on replay.As always, I'll start by saying that during the course of this call, we will make projections or other forward-looking statements that involve risks and uncertainties.
Our actual results and other future events may differ materially from those we describe here. These forward-looking statements are based on information, plans and estimates as of the date of this call, and Air Transport Services Group undertakes no obligation to update any forward-looking statements to reflect changes in underlying assumptions, factors, new information or other changes.
These factors include, but are not limited to, changes in market demand for our assets and services; our operating airline's ability to maintain on-time service and control costs; the cost and timing with respect to which we are able to purchase and modify aircraft to a cargo configuration; fluctuations in ATSG's traded share price and in interest rates, which may result in mark-to-market charges on certain financial instruments; the number, timing and scheduled routes of our aircraft deployments to customers; our ability to remain in compliance with our agreements with key customers and lenders; and other factors as contained from time to time in our filings with the SEC, including the Form 10-Q we will file this week.We will also refer to non-GAAP financial measures from continuing operations, including adjusted earnings, adjusted earnings per share, adjusted pretax earnings and adjusted EBITDA. Management believes these metrics are useful to investors in assessing ATSG's financial position and results.
These non-GAAP measures are not meant to be a substitute for our GAAP financials. We advise you to refer to reconciliations to GAAP measures, which are included in our earnings release and on our website.As Joe said, our third quarter results compare very favorably with our results from a year ago and show sequential improvement from results for this year's second quarter.
Our revenues again rose sharply, this time by $161.2 million or 79% from the prior year. Once again, a solid contribution from Omni was a principal positive factor.
Our work for the Department of Defense with contributions from all 3 of our airlines make it ATSG's largest customer, representing 34% of our revenues in the third quarter. 26% of revenues came from Amazon and 13% from DHL.Once again, our GAAP earnings include large unrealized effects of quarterly revaluations of several financial instruments.
These are mainly the effects of revaluing interest rate hedges on our bank debt, plus the noncash effects of changes in the value of the warrants we have issued to Amazon, which, this time, was positive due to a reduction in the value of our warrant liability. Those factors contributed to a tripling of our GAAP third quarter earnings from continuing operations to $105.1 million this year compared with $32.9 million a year ago.Diluted GAAP earnings per share for the third quarter were $0.19 versus $0.24 a year ago.
Warrant accounting excludes from GAAP earnings the per-share effects of mark-to-market changes in warrant liabilities when they are accretive. The adjusted earnings, EPS and adjusted EBITDA we reported this quarter exclude those warrant-related gains as well as amortization of warrant incentives for Amazon, affiliate losses reflecting our share of Airbus A321 development costs and the non-service-related costs of our retiree benefit plans.
As a result, our adjusted EPS for the third quarter was $0.31, as Joe mentioned a moment ago, versus $0.28 in the third quarter of 2018. Our third quarter adjusted EBITDA increased by $35 million to $109.2 million, and we are at $328 million through 9 months, up 52% over the first 9 months of 2018.Operating income for the quarter also rose 52% to $40.8 million.
Operating expenses increased to $325.3 million with significant increases in fuel, depreciation and amortization and salaries and benefits. The increases are primarily related to Omni and expanded operations for Amazon.
Quarterly interest expense increased from $5.6 million to $16.7 million, which includes the effect of higher borrowings for the Omni acquisition and aircraft investments.Our capital spending was $336.9 million through 9 months of 2019. We have acquired 9 Boeing 767 aircraft to date and placed 4 767s into service through September.
Our projection for full year CapEx is now $460 million, down $15 million from our August forecast. Joe will talk more about our fleet development plan shortly.You probably saw last night that we also completed another amendment to our senior credit facility.
The principal changes increased our revolver capacity by $105 million, consolidate our 2 term loans into 1 and reduced our interest rates at a given level of borrowing from the prior pricing schedule. We estimate that these favorable changes will cut our interest costs by approximately $4.7 million per year at our current borrowing level.On the operating side, our principal aircraft leasing and airline businesses continued to improve our bottom line as they expand their top line business volume.
ACMI Services on a pretax basis earned $4.4 million in the third quarter compared with a loss a year ago and earnings of $1 million in the prior quarter. We had advised you in August that third quarter ramp-up costs for expanded second half flight operations for Amazon would be about $4 million, which was roughly where we wound up.
We're now expecting another couple of millions of those costs in the fourth quarter as the updated flight schedules we received from Amazon in September require more crews than we had projected. That includes the effect of 2 more aircraft we will operate on a CMI basis for them.CAM, our leasing business, performed as expected.
Revenues net of warrant-related lease incentives increased 21% due mainly to 17 more aircraft in service versus a year ago, including 11 from Omni. CAM's pretax earnings were $17.4 million, down from a year ago and flat with the second quarter.
That's net of $4.8 million more in allocated interest expense and $7.7 million more for depreciation than a year ago, much of it related to the addition of Omni's owned passenger fleet. Costs and revenue shortfalls from delays in leasing newly converted aircraft and from aircraft transitioning between lessees were also factors.External customer revenues from other activities for the third quarter increased $4.6 million, primarily due to additional revenue for ground services and fuel sales.
That was offset by the termination of sort facility management services for the U.S. Postal Service in the third quarter of last year.
Pretax earnings from our other businesses were flat for the third quarter. The loss of business with the U.S.
Postal Service was offset by additional revenues to support Amazon's expansion, including gateway and aircraft fuel services and sort equipment maintenance.That completes the summary of our operations for the quarter. Joe?
Joe Hete
Thanks, Quint. ATSG looks substantially different today than it did a year ago.
On November 8 last year, we acquired Omni Air. Before the year ended, we won rights to acquire 20 more 767 feedstock aircraft and agreed with Amazon on leases for 10 more 767 [ph] freighters and extensions on 20 others we lease and operate for them.
Last spring, we added 3 years to our freighter leases and operating agreements with DHL.We are a bigger company now, and we'll keep growing over the next several years as we meet increased demand for our freighters and now the passenger 767s and 777s that Omni operates. We have also diversified our revenue stream with the Defense Department now representing 1/3 of our revenues.
Our growth is exceptional in comparison to the overall air cargo industry. Our operating fleet will reach 100 aircraft by the end of the year, up 19 from this date last year, and all our available freighters will be flying during peak.
We're growing because we principally serve the North American domestic express and integrated market, which is driven by e-commerce demand that is growing rapidly year-over-year. We're also growing because we have, in Omni, the passenger assets and reliable earnings stream from another market largely insulated from economic volatility, specifically transported military personnel.Omni's leadership in that arena continues.
As we look back on a year's worth of results from Omni since we acquired it, we can tell you that it has performed just as we expected and is poised to do well again in 2020. While others faced strong headwinds from slowing economies abroad and tariffs, our U.S.-based operations are moving forward.
The evidence of that trend is easy to find.DHL reports that its Americas market continues to outgrow its other global regions. UPS recently said that its second quarter overnight growth was up 30%, and third quarter was up 24% year-over-year.
E-commerce volume is being driven by the convenience of mobile technology and the integration of transaction platforms with social media. The projections we follow call for 20% growth in e-commerce worldwide and 15% growth in the U.S.A.
through 2021.Anticipating that trend, our 767 freighter growth pipeline is in good shape. We will convert and deliver to awaiting customers 8 to 10 767s next year, the majority of which are under agreement.
Our growth opportunities are primarily in the Americas but also in Southeast Asia and the Mideast. In addition to our agreement a year ago with Jetran for 20 ex-American Airlines 767s, we have been acquiring feedstock from other sources to meet specific customer requirements and market demand through 2021.
To date, we have purchased 3 of the 767s in our Jetran allotment. In short, we have largely locked up the feedstock we expect to require for the next 2 years.
This puts us in excellent position to service customer demand as the feedstock market remains tight.Even as we manage significant growth in our airline operations, we are leading the way in providing safe, reliable, on-time service to our customers every day. Both the words and actions of our customers affirm our performance quality.
Our freighter deliveries are backloaded this year as we told you would be the case. From late June through the end of the year, we will put 8 767 freighters into service for Amazon.
That's 6 that CAM is leasing them and 2 others Amazon has shifted to ATI from another carrier. We are also dry leasing 4 converted 767s to UPS, plus a fifth in January.
We expect to end 2019 with 5 767s, and we're awaiting modification.Given the short 2019 lease and operating periods for these new assets, they will deliver much greater cash flow in 2020 than in the fourth quarter when we will also absorb their start-up costs. So our guidance remains unchanged from the $450 million we gave you in August.
Our fleet growth profile for next year is shaping up to look much like this year as well: a full year of contributions from the aircraft we are placing now, plus cash flow, mainly in the second half from deliveries of 2020 aircraft.Quint mentioned that we have just amended our credit facility in ways that add flexibility and reduce our interest costs. That's a strong endorsement from our banks of our ability to generate exceptional cash flow to fund most of our growth while we continue to operate at comfortable or even decreasing debt leverage.
We're aware that many investors have turned away from the air cargo sector because of the macro factors we're all aware of. Certainly, our stock price is reflecting a lot of that skepticism.
Those who look closely at ATSG, however, will find a company with almost no fuel price or payload risk, reliable performance and in-demand assets. We are looking forward to years of strong cash flow without federal tax cash obligations generated from long-term lease and operating contracts.
Some of those leases extend late into the next decade.That's the performance we have been saying for years that our business model can and does deliver. We expect to moderate our fleet investments next year and do so further if e-commerce and other long-term trends driving our growth subside.
We would instead use more cash to pay down debt and return some to shareholders, but we have no intention of forfeiting today's growth opportunities to competitors because they drive the shareholder value we are generating today.That concludes our prepared remarks, John. We're ready for the first question.
Operator
[Operator Instructions] And our first question is from Jack Atkins from Stephens.
Unidentified Analyst
You've got Wade [ph] on for Jack. Congratulations on the quarter.
So to start, when we think about the puts and takes of your guidance, it sounds like there are some increased customer demand in the fourth quarter but also some ramp-up costs related to that and then some timing issues related to some aircraft deliveries. So first, are we reading that correctly?
And while this might mute some upside this year, does that provide some greater visibility into additional growth for next year?
Joe Hete
Well, I think the upside for peak season is just strictly related to peak in terms of demand for the assets that we normally have tied up in maintenance, et cetera, so we've got increased demand there in terms of its utilization. And as we mentioned, in addition to that, we've got 2 aircraft that weren't originally part of our plan that were transferred over from another operator that we had to gear up for.
So we're going to see costs associated with that. And then when you look forward, going into 2020, one of the things you run into is all training, for example, seasons for line crew members, other than the upgrade side in the fourth quarter, and then it picks back up again in the first.
So we continue to have to add people into the mix to accommodate the existing lift and then be able to transition into the first of the year to keep the aircraft going.
Unidentified Analyst
Okay, that makes sense. And so when we think about the goal for 10 deployments in FY '20 with 5 already under contract and sounds like strong demand for the other 5, could that plan for 10 move any higher?
Joe Hete
No. I think we're constricted by the production lines from a conversion standpoint, and that's why we usually say it's 8 to 10 because you can always run into an issue where you can't get the airplanes out as quickly as you would like to.
We pretty much occupy all the modification lines at IAI from a conversion standpoint. There's another airplane or 2 from other operators in there, but we pretty much monopolize them.
So the ability to kick that up is -- will be extremely difficult and most likely wouldn't happen.
Unidentified Analyst
Okay. And then last one, and I'll hand it off.
Can we talk preliminary CapEx outlook for 2020? Do you guys have that yet?
Quint Turner
Yes. We -- this is Quint.
We would foresee some reduction certainly from 2019's level, which we've guided for this year at $460 million. I think you would see us more closer to the $400 million type level, somewhere in that vicinity.
Obviously, we're not -- we haven't totally finalized all that, and we hope to give more guidance on that, of course, the next time we talk. But it is a -- we foresee a lower CapEx requirement certainly as we go into 2020.
Operator
Our next question is from Helane Becker from Cowen.
Conor Cunningham
It's actually Conor Cunningham on for Helene. Just on -- maybe coming at 2020 a little bit different, I mean, for the EBITDA outlook.
I know you don't -- you aren't giving formal guidance, and I think you spoke to a little bit today. But when we think about 2020, is it -- should we be thinking about it in the context of EBITDA growth of at least the growth in your fleet?
I mean just given what you did this year, I realize there are some start-up costs. Maybe you can give a little -- some puts and takes on what you're expecting maybe on the cost side and maybe on revenues if you have any visibility there.
Quint Turner
Yes, Conor. As you say, we, of course, will give the guidance when we -- as we always do in February when we come out with the end of the year.
But because of the timing of the aircraft entrants this year, you're right that full year contribution next year will be evident. We've got, I believe, 6 -- well, 8, counting the 2 that came in from Amazon from a different carrier, but we have 6 that CAM will be leasing that will be starting in the fourth quarter.
And so you're going to see a full year contribution from those. And we sort of -- I think -- the analysts that follow us kind of know roughly what we expect to see in terms of contributions from the lease for those as we look forward.And we've also had a lot of start-up costs in 2019 given the growth, the number of aircraft that we brought on.
So we would hope to see some efficiency there, although there'll be more new aircraft coming online next year. Certainly, it doesn't appear to be the number that we've had to gear up for this year.
We also had to gear up for a heavier flight schedule in the second half, and that was a pretty significant driver of the start-up costs that we've talked about this year. So having that behind us, we would expect to see some efficiencies in our airline ACMI operations next year once they've cleared those start-up costs.
So I do think that we'll have a tailwind on our EBITDA going into next year, and we expect a double-digit-type growth rate for next year.
Conor Cunningham
Great. And then just on maybe on the DoD business.
So how was that looking into 2020? Are you expecting it kind of to remain at a very robust level that you're seeing now?
Joe Hete
Yes. From a DoD perspective, we don't anticipate that we're going to see any reduction in that.
As the -- if you look out from a military perspective, as we said on previous calls, one of the things they're trying to do is minimize somewhat the utilization of the military's own assets, which means anything that can and could be done by the commercial sector will drive some of that. It's not going to be a significant growth number; but on a year-over-year basis, we expect to see growth.
Conor Cunningham
Great. And maybe one on the A321 conversion.
So you're targeting certification in mid-2020. When do you expect to start looking for feedstock for that aircraft?
And have you already started to have some discussions with customers on that plane? And does it actually open you up to additional customers that you don't actually have right now?
Just curious on your thoughts there.
Joe Hete
Yes. The 321, right now, the target is to get the certification, call it, in the latter part of the second quarter of next year.
The aircraft, from a conversion standpoint, is down off jacks. It's sitting on its own landing gear.
We're still doing some testing, expect to start test flying in the first part of the year going forward. As far as the aircraft itself, feedstock, we're always on the lookout for feedstock if the right opportunity would come along.
But right now, when you think about things like the 737 MAX being grounded, et cetera, a 321 is an equivalent lift, actually a little bit more, but that's going to hold feedstock availability down. But we're talking to a number of customers, both ourselves and Precision.
Obviously, we can leverage the 321 from the perspective of buying aircraft, converting them and leasing them out ourselves. We can leverage it from the standpoint that we're part owner of the ultimate STC that would come out, and so we'd share in those proceeds as well.Really, from a production standpoint, if you're talking about a second quarter certification of the aircraft, you might get a couple more aircraft into the mix in the latter part of 2020.
And from a full production standpoint, I don't see it being a big production line going forward until probably the latter part of '21 and '22.
Conor Cunningham
Okay. In terms of like the actual conversion, is it a similar time frame as the other aircraft that you normally do?
Or is there any, I don't know...
Joe Hete
Are you talking about the elapsed time to converting an aircraft?
Conor Cunningham
Yes.
Joe Hete
Yes. That's difficult to say at this point.
It is a fairly extensive mod, although you are actually obviously talking about a smaller airframe than the 767 is. But until we get through the certification process itself and understand all the piece parts.
As you can imagine, during the certification, there's a lot of form and fit issues that you have to deal with, but in no way shape or form should it be any longer than what the 767 is, which right now is, call it, 120 days.
Operator
Our next question is from David Ross from Stifel.
David Ross
It's actually me. A question on the -- yes, I guess on Omni first.
It's been almost exactly a year. So what's the post-deal postmortem in your analysis after having Omni as part of ATSG for the last year?
Is it better than expected, worse than expected? How is it different?
Quint Turner
Dave, it's Quint. And we are honored that you actually joined us on the call today, so.
Yes. It's a good -- great question, Omni.
As Joe said in his remarks earlier, it's performing as expected and I would say better in some respects certainly than expected. So we're-- as far as the postmortem, it was accretive from the beginning and has remained so to our earnings.
We talked about what our assumptions were when we purchased it, about its trailing 12-month EBITDA and cash flow production. And I would -- we've -- as you know, we've taken the 11 owned aircraft and they're part of CAM's portfolio.
So part of that EBITDA manifests itself with CAM. But in total, if you look at it, I think it is -- it has met and/or exceeded what our expectations were for cash flow production.
And moreover, in terms of, I think, putting us -- putting our revenue book in a place where it's less exposed to volatility from things like the general cargo market, it really was a game changer for us. And I think, given all the growth opportunities that remain with us in the e-commerce space, it's going to provide a nice balancing immunization impact on our exposures to the broad economy.
Joe Hete
David, if you look at it from a utilization standpoint, obviously mix comes into play if you're talking about 777s versus 767s. But if you look at third quarter, for example, on a year-over-year basis, Omni's block hours were up 16% versus third quarter last year.
So it's growing in addition to what we expected to get from a baseline perspective over the long term. And of course, as we've mentioned, I think, on the last call, we've given them an additional 767-300 out of the American fleet to operate in the -- not the military sector, but commercial sector.
And it's -- right now, as Quint mentioned, it's doing everything we expected it to.
David Ross
But the block hours are up mostly on an apples-to-apples basis, but there's one more plane in there?
Joe Hete
No, because the one more plane isn't going to drive a 16% increase in block hours on a year-over-year basis. If you look at that from a raw numbers perspective in terms of the -- if you look at it, there was 9,000 block hours flown by Omni in the third quarter of this year versus about 7,800 the prior year.
So you're not going to get that kind of utilization out of one aircraft. So everything in total is growing.
David Ross
Okay. And Omni did also bring some charter business to you, although it's not the long-haul cargo charter that's exposed to the air freight market and what's going on with trade and tariff issues?
How do you think about the charter business as part of ACMI? Does it change the profile of the earnings longer term?
Is it more volatile? Is there something about the passenger charter market that's more stable than the cargo charter market?
Joe Hete
No. I wouldn't say it's more stable.
I mean charter is what it is. I mean think about Omni in total.
Remember if you look at the book of business they had, it was 70%, DoD; 15%, government; and 15%, commercial. And one of the commercial -- on the commercial side, they had one long-term customer.
They had a dedicated aircraft to it. So it's not going to be a huge driver of revenue for us from a charter standpoint in terms of -- in comparison to the total book of revenue that we have.
But certainly, if you can put hours on an asset that otherwise might be sitting on the deck, it certainly helps the bottom line.
David Ross
And then, when we're looking at the fourth quarter and maintaining EBITDA guidance for the year, do you expect ACMI margins, given the puts and takes with the more aircraft, more hours but somewhat increased training costs, should ACMI margins be up year-over-year or flat or a little bit down?
Joe Hete
No. It should be up on a year-over-year basis, just because the number of additional assets we have flying out there.
But as Quint mentioned, we've incurred a significant amount of investment in training costs for crews, and that will roll into the fourth quarter as well. We've had about $4 million to this past quarter, third quarter, in terms of just training costs associated with bringing crews online.
And that will continue not to that degree in the fourth quarter, so margins should improve as we go through the fourth and start off the year.
Quint Turner
And remember, Dave, we only had Omni for a partial quarter a year ago, right? So that will have -- that will help margins having them for a full quarter this year.
David Ross
And last question, Quint, on the interest expense. I don't know if I was writing too fast or what.
But did you say that with the new credit agreement that interest expense is going to be down $4.7 million a year, so more like $15 million a quarter?
Quint Turner
Yes. At the current -- right.
At the current borrowing level, we essentially were able to achieve a lower pricing in terms of the spreads based on the pricing grid that the banks charge interest to us on. And so I think at our -- at current pricing level, immediately it would drop 37.5 basis points, which is you just take that across our secured debt at the current level, it equates to about $4.7 million per year of lower interest costs.
Operator
[Operator Instructions] And our next question is from Steve O'Hara from Sidoti & Company.
Stephen O'Hara
I guess just, first, on the start-up costs and things like that for the -- I assume it's just Amazon. But can you just -- I mean I think the total is something in the neighborhood of $13 million or something.
Is that kind of a -- I mean that should be a tailwind for next year. And then I assume this is kind of ramping up faster than expected.
But the way Amazon utilizes the aircraft, is that turning to sort of having extra crews maybe not as heavily utilized in 1Q and 2Q? Does that turn into more of a headwind having extra crews?
Or how does that work?
Joe Hete
From a crew costs or training standpoint, Steve, the number isn't quite as high as $13 million for the year on a forecasted basis. But as I said, we still will incur some in the fourth quarter as well.
As we transition into the next year, obviously, we have to again start people running through their recurrent training cycles, which they haven't had, as well as open up vacations again for folks. So you need people to backfill where otherwise they would -- they were not in a training mode, and you're gearing up for a regular utilization of the aircraft themselves.
So it will be a tailwind for us in the first part of the year. But then again, it all depends on what the demand is in terms of the flying.
We came into this year with -- expecting less than half of what we think we're going to incur this year from training costs, but that changed pretty quickly when the aircraft utilization changed, which primarily driven by Amazon is going to a 1-day delivery service.
Stephen O'Hara
Okay. All right, that's helpful.
And then in terms of the impact on the MAX grounding, do you know kind of roughly what that impact is maybe so far this year and then maybe what you expect it to be for full year, assuming that plane eventually starts flying again?
Joe Hete
Well, from the MAX standpoint, I'm not sure what you mean by the impact. Now we are flying one aircraft for Air Canada to cover for their MAX being grounded, so we're getting utilization out of one aircraft for that.
As I mentioned earlier, it does have some impact from a feedstock standpoint, particularly on a light kind of asset basis, the A321. There may be some folks out there that are holding on to a 767, for example, to cover for a MAX that may be grounded.
But outside of that, there's really no direct impact to us today, other than the fact that we are operating that one additional aircraft for Air Canada.
Stephen O'Hara
Okay. So I guess that was the impact I was talking about.
So it's really just one aircraft and probably not a huge driver of anything or negative when it goes away?
Joe Hete
No, it won't be.
Stephen O'Hara
Okay. And then maybe just lastly, in terms of the -- did you say kind of what the expected EBITDA run rate was at the end of the year?
Quint Turner
We didn't. We didn't a give a number for that, Steve.
And of course, we'll give that guidance in February. So I think you know it's certainly north of $460 million and less than $500 million.
How about that? So somewhere in between.
Operator
Our next question is from Chris Stathoulopoulos from Susquehanna.
Christopher Stathoulopoulos
I was wondering if you could talk about, given all the changes in ACMI over the last 2, 3 years with Amazon, certainly Omni, how we should think about pretax margins on a steady-state basis, assuming that there aren't any other significant acquisitions and perhaps in the kind of 2% to 3% GDP backdrop?
Quint Turner
Yes. I mean, Steve -- or Chris, we've always, I think, said that the ACMI Services margin is a single-digit-type margin.
It has more volatility associated with it because the changes in customer demand, such as we've had this year where you're ramping up operations, can impact it. Omni, if you look back at our historic ACMI margins, you'll note that there's been improvement, certainly, with the addition of Omni and with their customer base and the nature of their contracts.
So I still believe we are looking longer term at a single-digit margin target for that. It's -- certainly Omni has helped move that a little higher, if that helps.
Christopher Stathoulopoulos
Okay, that's helpful. In the release, you talked about lower CapEx in the out-years and reducing leverage.
And Joe in his prepared remarks said that you have visibility into your feedstock, I think, for the next 2, 3 years. So putting all that all together, are you suggesting that beyond 2020 or so we could see free cash flow, positive free cash flow on a sustained basis, which I know is kind of contrary to the way the model works with an aircraft to grow, et cetera?
But I was just curious how to reconcile that versus what you called out in the release.
Quint Turner
I think that certainly we've been in an investment cycle given the growth opportunities that were in front of us. And as Joe said, when good growth opportunities are there, we like to pursue those because that produces the best value long term.
I do think that we're in a great position in that our maintenance CapEx needs are pretty low. We said only about $100 million.
We were proactive in securing feedstock when we bought the 20 aircraft fleet, ex-American fleet, and it's fit well into this investment cycle. But if you look beyond that, I mean, I think you'll see plenty of opportunity for free cash flows, deleveraging and free cash flows.
And there's a lot of capital allocation option certainly that we'll have at our disposal if you look out. Really, you don't even have to look that far for that.
Christopher Stathoulopoulos
Okay. And then, just one follow-up here.
So how dynamic the peak season has been over the last few years, looking specifically at the integrators and all the changes, investments they're making in their network to handle volumes. I'm curious, as a lessor, you operate some of these planes, how you've seen the peak shape?
What's perhaps easier today as an operator? Or more challenging, how are you balancing utilization and on-time performance for a business that continues to grow in the double digits?
Quint Turner
Chris, Mike Berger happens to be sitting here with us. And since, Rich Corrado usually fields those kind of questions, but he's traveling.
We're going to bring Mike, our Chief Commercial Officer, into play here. So Mike?
Michael Berger
We'll have all our aircraft flying during peak this year, and we've been committed to that for a number of weeks now. So we saw extremely strong demand early on from some of our customers, so we'll get full utilization out of our fleet this year.
It's also a little bit more condensed this year with Thanksgiving falling in the last week of November. So that will make crunch time, so to speak, even more important in terms of on-time performance.
But as Joe stated in his remarks -- earlier remarks, we continue to see all of our airlines performing at very high, on-time performance levels. So peak this year, shorter in terms of a condensed standpoint and full utilization of all of our aircraft in our fleet, so very good season for us.
Joe Hete
One of the challenges going forward, Chris, from the industry's perspective is there's very few assets that aren't already dedicated to utilization on a 12-month-a-year basis. And when you start talking about peak assets, you have to have something that either has some book of business that occupies it for 10.5 or 11 months out of the year and then frees it up, which those are -- just really aren't out there to any great degree at all.
So it's going to be a challenge for the industry as a whole from a service perspective to deal with peak volumes as the e-commerce sector of the business continues to grow and stuff has to move by air versus service movements.
Christopher Stathoulopoulos
Okay. Last one, you say Rich is overseas, and, Joe, you talked about e-commerce potential overseas.
Outside of Amazon flying in, say, Europe, I'm just curious what other e-tailers or operators are out there that you can point to that might look to supplement their existing air capacity or perhaps start up a dedicated fleet?
Joe Hete
Yes. I mean there's not a lot out there in terms of that have the bandwidth to.
I mean people always talk about Walmart, for example, but Walmart has such an expansive surface distribution network that it'd be difficult. And as far as I know, they don't do a darn thing by air today.
I think you have to start looking around the globe to find other opportunities. Certainly, Amazon, as you mentioned, in Europe could be one.
You could get into Asia, for example, but our ability to operate there is pretty limited. But certainly, it would be a potential outlet for leasing aircraft in there, especially when you start talking about something like the 321.
One of the reasons we like that airplane is great cubic capacity, much better than the 737-800. So we see potential markets in the long pool if, for example, an Alibaba or something like that sets up a network in the Asian sector.
That would certainly be a great location for the 321 because, right now, a bunch of the things that they do in China, for example, is point-to-point as opposed to hub-and-spoke.
Operator
Our next question is from Christopher Hillary from Roubaix.
Christopher Hillary
Just a quick question. I know there's a lot of differences, but do you have any observations on the acquisition of Aircastle?
As earlier you kind of commented on how the stock price might have reacted to some negative trends in the freight market? I was curious if you thought there was any read-acrosses [ph] from what happened with Aircastle.
Joe Hete
No. Nothing we can see.
I mean generally your larger leasing companies are focused more on new -- buying new assets, primarily focused on the passenger market. We're -- as we're on converted assets in the cargo market, so I don't think there's much overlap there.
Operator
We did have another question queue up from Chris Stathoulopoulos from Susquehanna.
Christopher Stathoulopoulos
Just on -- I don't think anyone's asked about labor but the open CBA with one of the subsidiaries. I think it's ATI.
And then also, given all the aircraft additions for this year and next year, how we should think about headcount and maybe if you could split that between pilot and crew versus corporate.
Joe Hete
Well, I think from -- it's actually ABX that has the open collective bargaining agreement, not ATI. They're still under the auspices of the National Mediation Board.
They met for 3 or 4 days last week with the federal mediator, so negotiations continue. Now that we're heading into peak season, you probably won't see any additional mediated discussions until after the -- after we get through the first of the year.
But like I said, from our perspective, even though we've got this collective bargaining agreement that's open, the crews continue to move the airplanes on an on-time basis. Their performance remains at the usual high level.
So things are on an even keel there. And second part of your question, Chris, was?
Christopher Stathoulopoulos
Just with all the aircraft additions, how we should think about headcount growth for 2020 and if you could maybe split that between pilot and crew versus corporate.
Joe Hete
Well, it's difficult to say until we know what the schedules would be from a pilot standpoint. Like I said, 4 of the aircraft go to Amazon, 1 to UPS and -- of the 5 that we have firm commits on.
As they add aircraft into the network, it could lower the utilization on other assets as they add those in. So it's difficult to say what the actual number would be until we receive schedules for that.
As far as the -- let's call it, the admin side of the equation, it's a pretty small number overall that you have to add in order to support the growth of the airlines themselves. Probably the bigger driver is on the maintenance side of the equation as you have to add technicians depending upon whether you're using an existing location to run that airplane through or a new location.
And all the talk you hear out there in the industry about pilot shortages, I'm here to tell you that I think there's a much more acute shortage of qualified maintenance technicians than there is pilots. And of course, if we can't fix the airplane, you don't need a pilot to fly it.
So we've had no difficulty attracting pilots to date with ATI because of the great opportunities, so it's -- we're able to put butts in the seats, so to speak.
Christopher Stathoulopoulos
Okay. And then last one on the Amazon warrants; have any been exercised?
And just remind us again when the 2016 and 2018 tranches expire.
Quint Turner
Yes, Chris, nothing exercised as of yet. The first warrant tranche, which we issued in, what, March of 2016, was a 5-year exercise period, so March of 2021.
And then last December, we issued the other tranche associated with the lease extensions on the 20 aircraft, the additional 10 leases in the CMI agreement, 5-year extension. Those warrants have an exercise period of 7 years from that date.
So that would be, what, 2025, late 2025.
Operator
And we have no further questions at this time.
Joe Hete
Well, thanks, John. On behalf of the Board of Directors, I want to thank those of you on the call today who shared their thoughts last month about the factors affecting our stock price.
We appreciate all the ideas about how we can make ATSG a stronger, more profitable and more shareholder-focused company. Thank you for your continuing support, and have a quality day.
Operator
Thank you, ladies and gentlemen. That concludes today's call.
Thank you for participating, and you may now disconnect.