Good afternoon. My name is Thea, and I will be the conference operator today. At this time, I would like to welcome everyone to the Apache Corporation fourth quarter 2017 earnings conference call. All lines have been placed on mute to prevent any background noise. After the speakers' remarks, there will be a question-and-answer session. Thank you. At this time, I would like to turn the conference over to Gary Clark. Please go ahead.

Good afternoon and thank you for joining us on Apache Corporation's fourth quarter 2017 financial and operational results conference call. Speakers making prepared remarks on today's call will be Apache's CEO and President, John Christmann; Executive Vice President of Operations Support, Tim Sullivan; and Executive Vice President and CFO, Steve Riney. In conjunction with this morning's press release, I hope you have had the opportunity to review our fourth quarter financial and operational supplement, which can be found on our Investor Relations website at investor.apachecorp.com. On today's conference call we may discuss certain non-GAAP financial measures. A reconciliation of the differences between these non-GAAP financial measures and the most directly comparable GAAP financial measures can be found in the supplemental information provided on our website. Consistent with previous reporting practices, adjusted production numbers cited in today's call are adjusted to exclude noncontrolling interest in Egypt and Egypt tax barrels. Also please note that with our exit from Canada, forward guidance and future quarterly reporting will refer to the United States or the U.S., and we will no longer use the term North America. Finally, I'd like to remind everyone that today's discussions will contain forward-looking estimates and assumptions based on our current views and reasonable expectations. However, a number of factors could cause actual results to differ materially from what we discuss today. A full disclaimer is located with the supplemental data on our website. Our prepared remarks will run a bit longer today, as we have a lot of information to cover with fourth quarter and full year results, as well as our three-year outlook. So we will extend this call past 2 o'clock in order to accommodate 30 minutes of Q&A. And I will now turn the call over to John.

Good afternoon, and thank you for joining us. On today's call, I will begin by discussing Apache's approach to the current environment. Then I will highlight our 2017 accomplishments and provide an overview of the fourth quarter before turning it over to Tim and Steve for more details. And finally, I will close with commentary on our 2018 to 2020 outlook. Early in 2015, we focused Apache on a path that creates real value for our shareholders over the long term and through the commodity cycles. The foundation was underpinned by cost and capital discipline. And we linked our compensation plans accordingly. This approach has driven our actions through a very challenging period. Specifically, we have streamlined our portfolio, reduced capital investment, increased capital productivity and efficiency, and reset the corporate and operating cost structure. By choice, we allowed production to decline rather than chase growth in an environment that was mostly value destructive. This disciplined approach enabled us to cost effectively lease and quietly discover a world-class unconventional resource of scale and to direct our limited additional investments to improving long-term returns. We were also able to maintain our dividend, while many in the industry were reduced or eliminated. Moreover, we strengthened our balance sheet by reducing debt and avoided diluting our shareholders' long-term value by electing not to issue equity to pursue expensive acreage acquisitions or to fund an outsized capital program. We have been clear in our belief that an E&P company, over the long term and through the commodity cycles, must be able to do the three things that leading companies in mature industries do, live within cash flows, grow the enterprise through prudent investment, and return capital to shareholders through a competitive dividend and/or share repurchases. This is the path that Apache is on now.…

Thank you, John. My remarks today will include operational activity in key wells in our U.S. and international regions, planned activity levels by play for 2018, and I'll conclude with commentary about the use of new technology throughout the organization. Operationally, we had a good year and continue to improve in key areas. Our fourth quarter production results continue the growth trajectory established in the third quarter. In the U.S., fourth quarter production averaged 222,000 barrels of oil equivalent per day, up 7% from the third quarter. U.S. oil production increased to 98,000 barrels of oil per day, an 8% increase from the preceding period. Much of this growth was driven by continued success in the Midland Basin. At the Powell field in Upton County, we brought online 20 wells from three pads with an average 30-day peak IP of nearly 1,400 BOE per day. These wells were drilled to the Wolfcamp B formation and are comprised of 1.5- and 2-mile laterals, producing in excess of 75% oil. These are half-section spacing and pattern tests, designed to methodically and scientifically determine the optimal development plan for the area. We continued to make good progress on drilling, completion, and cost optimization, as we use capital more efficiently with larger well pads and longer laterals. In the Midland Basin, we have reduced drilling and completion costs 20% over the last 12 months on a treated lateral foot basis, while production volumes improved 17%. These results give us confidence that we can continue to improve here and elsewhere in the Permian Basin. As John mentioned, we are moving to more pad operations at Alpine High. The Dogwood State pad is a six-well spacing test located in the northern flank and was selected due to retention requirements. These are dry gas wells on 660-foot…

Thank you, Tim. Today I will highlight Apache's fourth quarter and full year 2017 financial results, discuss our 2018 outlook, comment on cash returns and return on capital employed, update our Alpine High midstream progress, and briefly review our hedge positions. Before I get to these details, let me first review a few of the company's key financial achievements in 2017. Apache returned to profitability in 2017, both on a GAAP reported basis and on an adjusted earnings basis. We reduced our absolute level of debt and our net debt, ending the year with more cash on hand than we began with. We retained our investment-grade credit rating. We returned nearly $400 million of capital to shareholders through the dividend. And asset sales generated $1.4 billion of proceeds and eliminated approximately $800 million of future asset retirement obligations. In 2018, we will continue to maintain a strong balance sheet, direct our capital investments for value and long-term returns, and take important next steps to progress Alpine High into full operational mode, so we can return to living within cash flows as we prefer. Through all of this, we will also continue to return capital to our shareholders, which is an underappreciated aspect of Apache. Over the last three years, we've returned over $1.1 billion to shareholders through the dividend. For the next three years, we plan to return at least this amount, and possibly more, through the dividend or through share buybacks. Turning to our fourth quarter and full-year results, as noted in our press release this morning, under Generally Accepted Accounting Principles, Apache reported fourth quarter 2017 net income of $456 million or $1.19 per diluted common share. Results for the quarter include a number of items that are outside of core earnings and are typically excluded by the…

Thank you, Steve. Before we move on to Q&A, I'd like to comment on the 2018 to 2020 outlook we provided in this morning's press release. Over the next three years, we plan to invest a total of approximately $7.5 billion in the upstream, with just under $2.5 billion budgeted for 2018 and increasing slightly through to 2020. Additionally, we expect to invest $1 billion in the midstream build-out at Alpine High over the next three years. This will include around $500 million in 2018 and another $500 million split evenly between 2019 and 2020. As we have made clear, we are exploring funding alternatives for our midstream assets and are working to eliminate some or all of this capital from our three-year plan. I cannot overstate the strategic importance of the midstream solution at Alpine High. The optimal outcome requires a deliberate and thoughtful approach, highly integrated with the upstream development plan, and we are investing the necessary time and resources to get it right. The outcome of this capital program is a projected compound annual growth rate of 11% to 13% for Apache as a whole and 19% to 22% in the U.S. over the next three years, which will be accompanied by very solid returns. This growth will be driven almost entirely by the Permian, which we project to grow at a compound rate of 26% to 28%. Our international operations in Egypt and the North Sea will continue to be free cash contributors. For the last two years, we have been investing below our $700 million to $900 million maintenance capital rate. As a result, we have seen international production volumes decline. And for now, we anticipate they will continue on a shallow decline rate, given our planned investment levels. In the North Sea, however, we…

We'll pause for just a moment. The first question will come from John Herrlin with Société Générale. John P. Herrlin - Société Générale: Yes. Thanks. Regarding your Midland well pads, John, how many wells per pad? And is the 1.5 to 2-mile length ideal? Is that your optimal horizontal length for those wells?

Yeah, John. Thank you. First off, we brought on three half-section pads. And a lot of those were 8 and 10 wells, so we're testing multiple zones and the patterns there, and it's really important. I mean we'll be watching those wells now over time. We've got a couple of other – we've got a pad coming on in the first quarter, as we said, of 2018. And some others planned in 2018, which will help us with that. Right now, 1.5-to-2-mile is probably optimal. And a lot of that hinges on your land position. So we've had to do some work to kind of be able to block up some trades, to be able to drill those longer laterals. But I do think that the 1.5-to-2-mile laterals are going to be optimal for now. And obviously we're watching the spacing. I think that's one of the big keys is getting this spacing right. As you know, you can't take back wells you drill. And we've seen some instances where others have plowed ahead and are over-drilling. And now you're seeing a lot of interference. And so I think it's important to take the time, effort, collect the data, and do the science to make sure you get those patterns and spacing tests right to maximize the long-term returns. John P. Herrlin - Société Générale: With respect to the data gathering and all that, how much incrementally or how much has that been for the well costs or the pad development costs, the technology that you're focused on?

Well, we had a pad where we ran fiber on all the wells. We ended up spending a couple million dollars there ultimately, because of how we collected it. We also had seismic crews out there. And we actually collected true seismic data in between each stage. So we've done a lot of things, John, where we've invested that money. But when you look at the grand scheme of things, that relative to just one well that you over-drill is money well spent. So I mean it's worth taking the time and making sure we do this right. John P. Herrlin - Société Générale: Great. That's really it for me. Thanks.

Thank you.

The next question will come from Bob Brackett with Bernstein. Robert Alan Brackett - Sanford C. Bernstein & Co. LLC: Hi, guys. Question on the U.S. non-Permian, non-Alpine High. If I kind of take your guidance and back out what it implies for U.S. non-Permian, it's kind of under-investment, like a decline. If I look at the Permian outside of Alpine High, it looks like you guys basically keep it flat through 2020. Is that about right?

When you look at the capital program with where we are today, Bob, yes. We are spending very little outside the non-Permian capital. I think you're actually going to see though, that it's – with the investment that's come off of the last three years, it's not going to decline a whole lot. In fact, our Mid-Continent stuff will actually grow with just the five wells we brought on in the SCOOP this year. So we're in a pretty good spot there. But, yes, very little capital there for 2018. Now as we get out past a couple of years and we start generating a lot of free cash flow from Alpine High, we see that changing quickly. And so we like having the optionality there in those assets. And right now in the other Permian, it's close, a little more than maintenance. But it's really more designed at going at the right pace there. As I mentioned too in the Midland Basin, I mean we've shown – if you look at the results, the back half of 2017, as I mentioned in the script, we peaked in the fourth quarter of 2015. And then we really shut the programs down, reset the cost structure. Bottomed second quarter of last year. And then quickly in a matter of two quarters made a new production high. So really it's an off/on switch. It takes a couple of quarters. But it shows you the quality and the progress that we've made in terms of being able to apply that capital to the other Permian assets. Robert Alan Brackett - Sanford C. Bernstein & Co. LLC: And if you have those assets that other people would voraciously drill, at what point do you say that those belong in somebody else's hands? Or to your point, do you keep them for future optionality?

Well, we just have to weigh the value of that. I mean we look at the portfolio. We work the portfolio very hard. Last year we made a strategic decision to exit Canada, which we're very glad we did. I think one of the things that gets lost in there is that we eliminated $800 million of ARO, but that was a big strategic decision for us. We also unloaded some acreage that we felt like we got some very – prices for. And is exactly that. Something that somebody placed a high enough premium on that we felt like it would be better in their hands. So you're always looking at those things and weighing those things. And we continue to do that in the future. Robert Alan Brackett - Sanford C. Bernstein & Co. LLC: Okay, thanks.

Thank you. [Technical Difficulty] (59:22 – 59:31)

Jeffrey, your line is open.

Thanks. On the multi-year plan, you mentioned the 10% service cost inflation. Can you talk through some of the other assumptions, specifically what commodity prices you're underwriting? And if you're including any expected efficiencies or productivity improvements?

On the service side, yes. As Tim said, we have most of our big ticket items – rigs, frac crews, sand – under contract and tied up for the foreseeable future. So we feel good about the main services. But we are seeing the smaller things that drive the day to day, trucking, simple as the backhoes, pads. Everybody is wanting to raise costs everywhere. So we did bake in, in general, a 10% rise. Now when you look at some efficiencies at – and then we kind of took each play, play by play. As Tim told you, in 2017 we were able to reduce our Midland Basin well cost by 20% on a treated lateral foot basis and increase productivity by 17%. So we've taken those kind of play by play into account. The greatest efficiencies we'll see at Alpine High is we're early in that play and really starting to move into pads. And then less in the more mature plays, where we've drilled more wells. And so we've got a pretty conservative forecast on the capital side going into this year. And what I don't want to be is in a couple quarters, having to raise my capital, because I assumed that we could keep things down when the reality is there's a lot of pressure out there on a lot of different fronts.

I appreciate that detail. And then I guess on the commodity price assumptions through the three-year timeframe. Just trying to get a good sense of what's under-written, when Steve talked about cash flow neutrality and the upstream piece this year, for example, and just how that might evolve over time.

Yeah, Jeff. So obviously, there's lots of conversation out there today about cash flow neutrality and pricing assumptions and what pricing assumptions people ought to use. And obviously I think that the last few years have demonstrated how important we believe cash flow neutrality is. And we've said that many, many times that we ought to be able to live within our means. I think it's important that we actually acknowledge there are lots of different definitions out there about cash flow neutrality. There are lots of different methods that people use to talk about that. Some include dividends, some don't include dividends. Some actually go so far as to include asset sales. And some have some capital structure changes, all contributing to cash flow neutrality. Just to be clear, we take a very, maybe extremely pure approach, because we believe cash flow neutrality means that with no asset sales and with no changes to debt or equity, that you should end the year with the same amount of cash on hand that you began the year with. And that's a very pure definition. I'm not sure there's a more accurate definition of cash flow neutrality. If there is, I'd like to know what that is. So with that said, our plan for 2018 and beyond, what we talked about today around – or I talked about around 2018, is so the midstream is obviously operating at an out-spend, about a $500 million deficit. That doesn't – that's regardless of what price assumption that you might use. And there's obviously – there's some reasons that some of that might go away. And what I talked about is that the upstream, which is everything else in Apache, including dividends, that that would be cash flow neutral at current pricing is what…

Thank you.

Yeah, all of that, just to be clear, again, that all includes the dividend.

Understood. Appreciate all the color there as well. Thanks.

The next question is from Charles Meade with Johnson Rice. Charles A. Meade - Johnson Rice & Co. LLC: Good afternoon, John, to you and your whole team there.

Good afternoon, Charles. Charles A. Meade - Johnson Rice & Co. LLC: Thank you, John. I want to go back to something that you spent some time on in your prepared comments. And I'm hoping you can perhaps add a little bit more. And that's the discussion around the – said maybe the balance of returns between your Alpine High and say your Midland asset. And it sounded like you were saying in the short term, perhaps a better return is available in the Midland Basin. But if you widen the perspective a bit and look at the longer term trajectory in Alpine High, that actually that's going to deliver better returns to Apache shareholders. I wonder is that the right way to interpret your comments? And if so what is the timeframe where maybe there's a breakover that Alpine...

Well, first thing is the returns in both programs are excellent, so it's really not a return difference thing, Charles. The point was we could increase short-term oil by going much quicker in the Midland Basin today or some of the other Delaware Basin stuff today. The point was in the Midland and the Delaware, we're gathering a bunch of data by moving to the pads and the pattern spacing tests. And that's really, really important data that we're collecting right now. And quite frankly, you want to – the market has gotten conditioned to thinking that early performance in IPs is a direct correlation to EURs, which is just not the case. You have to look at how these wells perform and how these pads perform over a longer time, and especially as you start to look at the inside wells and so forth. So in our Midland program, we've actually brought on three brand new half-section pads late last year, which we're going to critically watch. And they're a little different configurations and we collected a lot of data. So there are two elements there. What my point was is we could accelerate the short-term oil over the longer-term investment at Alpine High. At Alpine High, you have a totally different animal, though. You have 6,000 feet of hydrocarbon column. We've got a 70-mile fairway, 340,000 acres that we control, multiple zones. We've now proven over 11 different landing zones across just the vertical column, and there's many, many more as we work through that. So it takes time. We're moving the infrastructure forward. And most of that's geared to the wet gas infrastructure that we have to have to process that and get put in place. And so when we look at advancing that over time and then you just look at the velocity at which we'll be able to reinvest that capital because of the F&D and because of the turnover and the returns, from a longer-term perspective, what's in our best interest now is advancing the Alpine High at this pace and the Midland Basin at the pace we're at. Charles A. Meade - Johnson Rice & Co. LLC: Okay, got it. I think that makes sense, John. And then this is more just a rifle shot question. The Dogwood and Elbert State pads, your Alpine High pads that you gave us some results on, I believe you gave us the oil cut. But what was the – if you could, give us an idea of the NGL yield on those.

No, the Dogwood pad we said is in the dry gas. I think the thing that's very exciting about that is that's a six-well pad that's on 660-foot spacing. And quite frankly, that's a heck of a lot tighter than what our location counts would be today. So the performance there is very encouraging. And it could lead to an increase in numbers of well counts, because it's performing very, very nicely. But the Dogwood is in the dry gas. And as we mentioned, we've got six wells on. They're 660-foot spacing, and they're currently making over 75 million [cubic feet] a day, and have already made 2 Bcf in a very short term while they were cleaning up. So it's really the first true pad and pattern. And it's validating the organics. This is a true shale. And so you're seeing great response from the first pad. The Elberts are shallower. And there you see the liquid yield goes up. Those are still cleaning up as well. It's a two-well pad. I think the key there was we got the cost down. And what you're seeing there is the liquid yield. And that will be wet gas too as it moves up. So I don't know, Tim, do you have the NGL yield on the Elberts?

This is high BTU gas. And under the cryo, it's going to be in the 140 barrel per million range of a typical wet gas well.

And it's about 60 barrels a million right now of oil too, correct?

Correct. Charles A. Meade - Johnson Rice & Co. LLC: Great, that's what I was looking for. Thanks, Tim, and thanks, John.

Thank you, Charles.

The next question will come from Michael Hall with Heikkinen Energy.

Thanks. I guess first on the trajectory of Alpine High in the three-year outlook, I'm just curious. Should we think about that as basically filling up infrastructure as you go, or do you have remaining infrastructure capacity to continue growth beyond that timeframe?

No, Michael, it's a pretty conservative plan. It's really based on retention, half the capital going to retention. We will be ahead of that. I said we would end the year at about 830 million a day of inlet processing capacity at the end of this year. And the volumes we have on that outlook don't fill that up. So we've got a lot of capacity there. And quite frankly, I think there's room for the picture to improve greatly as time marches on this year, both in terms of the volumes as well as the liquid content and the oil content as we test more of the zones and go forward there.

And in what timeframe does the Alpine High project as a whole go from being in an investment phase where there's an outspend, including the midstream, to – let's just set aside monetization, but to a free cash flow phase where you're actually harvesting the cash flows from the program?

We look at it – probably the best way to put this to you is on a rig line basis. A single rig line is going to turn cash flow positive in less than two years. And so if we hold the rig lines constant, you'll see it turn pretty quickly. So that's how we think about it. We laid out the capital on the midstream was (1:14:04) $500 million last year. We said $500 million this year and then $250 million, $250 million. So it starts scaling down. So then it really comes down to the pace on those rig lines. But you're going to see this thing start throwing off a tremendous amount of cash in less than a two-year window per rig line.

Okay. And then last on my end is just around the cost structure. I was just looking year on year at the LOE cost guide. It's up despite adding a pretty large wedge of what I would think is pretty low-cost gas. What's going on there that's not driving a reduction in LOE per unit? And should we anticipate a reduction over the longer timeframe within the three-year outlook?

There's no doubt that, as we mentioned in there, the per-barrel numbers are going to come down on Alpine High. Some of that's with starting to put various things in the LOE lines that hadn't been there prior. Some of it is when we pored our plan, our guys have taken a pretty conservative approach for what LOE looks like right now just because of the pressure we're getting for everywhere to raise the small things. So I think there's room, as we have historically done, to work on those numbers and beat those numbers. But it's probably just a little bit with this price environment changing so dynamically, that and the timing of some pads and some things coming on that is driving that. But there's no doubt over time the LOE per BOE for the Permian is going to come down significantly as Alpine High ramps up.

And at the corporate level, should we see that by 2019 you think, or more time?

Yes, this is Steve. So I think that a lot of that will depend on what happens with the midstream at Alpine High. Because John made reference to what we believe the LOE per BOE at Alpine High will be in his prepared remarks. That was excluding the midstream costs. And today, all of the midstream costs show up, or the operating costs for that show up as LOE. And obviously, as you're building that out, you're building out capacity and starting it up and operating it below its actual true capacity, in some cases and for some periods of time well below its true capacity. You're going to have a pretty high cost per BOE going through that. Two things will change that, number one, just ramping it up to a larger scale and really getting it ramped up to efficient activity. But also eventually what's going to happen, or at least I would anticipate will happen, is that the midstream assets become part of a midstream enterprise separate from Apache. And depending on the accounting treatment of that, the structure of it and control and ownership, you could see all those costs move over to gathering and transportation, which will be more of a typical third-party type of transport and processing fee as opposed to LOE. But that's going to have a meaningful impact, especially on 2018, the startup and operations of that midstream enterprise.

All right. Look forward to seeing how it all progresses. Thank you.

The next question is from Brian Singer with Goldman Sachs. Brian Singer - Goldman Sachs & Co. LLC: Good afternoon.

Hey, Brian. Brian Singer - Goldman Sachs & Co. LLC: With the three-year outlook on Alpine High to get to 160,000 to 180,000 BOE a day, fully recognizing that production mix is different from value, can you just give us the update on your expectations for the oil versus NGLs versus gas split in 2020? And then on the gas piece, what are your expectations for how much will go beyond the local market in 2020? I think you referenced the Gulf Coast Express Pipeline as one option, but maybe you could quantify that a little bit more.

in the prepared remarks, we said fourth quarter, you were 83% gas, 10% NGLs, and 7% oil. And we said in my prepared remarks by 2020 that the oil, we assume the oil will remain about constant. And the NGL volume's going to grow to about 30%. So you're going to be more like 63% gas, that's probably then going to be two-thirds to – or more wet gas to dry gas with a heavy 30% NGL. And then we've assumed a 7% oil mix. But I believe that's pretty conservative. And that's – as I mention right now, our plan is geared more towards retention and what we see without the ability to be able to drill more at the other locations. So that's how that'll transition. And...

And on the marketing side, we've got a lot of activities actually underway here, a lot of parts that we're working on. We're continuing to contract gas at or around Waha or near our assets. We're continuing to work on contracting gas down on the Gulf Coast, where we now have the capacity to transport 500 million a day starting in 2019. So we're working on marketing contracts at the Gulf Coast from that point forward. And then we're continuing to look at more gas opportunities as well as – for the longer term, as well as oil and NGLs. And so I think there's still a lot of work to do on the marketing side, both physically moving product and also selling that product or downstream products from those products. Brian Singer - Goldman Sachs & Co. LLC: Great. Thank you. And then shifting to the other side of the world, there was some news earlier this week of some contracts signed to move natural gas into Egypt in a couple years out, at a price ostensibly greater than what Apache receives for its gas. I realize that Apache's price is probably a blend of a number of contracts and concessions. But how prominent are the opportunities to grow gas production on your concessions? And how interested would you be in investing or monetizing that in some other way?

I mean there are definitely opportunities to move that forward. I mean I think we just added 40% to our acreage footprint. Most of our revenue comes from the oil side, quite frankly. And the way a lot of our concessions are structured and so forth, some of those prices are set. The Egyptian government has been very flexible and willing to step in and structure things that would encourage some development of some different types of things. So I think you'll see us continue to do that. And if it makes economic sense in relation to the oil that we're developing over there, we will do that. In general, we applaud the contracts to bring the gas in. I think it's a good thing for Egypt and the country and actually, it's very beneficial to us. So we're happy to see what's happening, both in the deepwater with the gas developments there as well as bringing in the gas that was recently announced. Brian Singer - Goldman Sachs & Co. LLC: Great, thank you.

The next question is from Leo Mariano with MetAlliance (1:21:37).

Yes, hey guys. Just in terms of the Alpine High midstream monetization, recognize that it sounds like you guys are still contemplating various avenues here. But just try and think of it from a high level perspective. Is this likely to be an event you think that's going to come later in 2018, or more of a 2019 event?

It could come any time in 2018 I would guess. I'll be honest with you. We made a lot of progress. We've had some inbound proposals that are fairly attractive. There's a tremendous amount of interest. And quite frankly, I think there's a lot of folks out there that realize this is going to be one of the most critical pieces of infrastructure in the Delaware Basin. And so I'm very optimistic that we'll be able to get something done and something that'll be very, very strategic for Apache.

All right, that's very helpful. And then just looking at kind of the Alpine High project and thinking about some of the splits that you threw out there in terms of hydrocarbon mix. You certainly talked about strong rates of return for your shareholders here. Just wanted to kind of think about some sensitivities there. Have you guys looked at kind of some of the downside cases, where if say you're getting $2 for the gas over the next three years and $20 for the NGLs, does the project still have kind of the right hurdle rates for Apache?

It – the beauty of it is, is with the wet gas, you don't need the gas. Now you got a scenario where you have really low NGL prices and really – I mean obviously, as under a scenario where all commodity prices go way down, then it's a different story. But this thing's going to really hum below $2 on the gas side. And I think with what's going on on the Gulf Coast, with the expansion that's taken place in the petrochemical end, we look out to 2019, 2020, 2021, and we see a pretty robust NGL market, as well as the ability to get the gas to the Gulf Coast. So it's what sets this play apart is the cost structure. And ultimately, it's the cost to drill the wells that's going to be superior and the deliverability. And it's that combination with the liquid yields and the oil production is what makes it unique. And quite frankly that's what we know makes it differential. And the other factor is you're not going to have to move a lot of water in the lower zones. And that's another very differentiating fact. But, yes, we've run many cases on the downside. We would not be making this type of investment on the midstream or the upstream side if we thought there was a sensitivity that was close to anything that would come into making it not work under very, very low gas and NGL and oil prices.

Thanks, guys.

The final question is from Doug Leggate with Bank of America Merrill Lynch.

Oh, hi. Good afternoon, everybody. Thanks for squeezing me in. John, you've given a fairly robust defense of the Alpine High in terms of the returns, the cost structure, and so on. And certainly, the production growth looks pretty impressive. Can you give us an idea of what you anticipate the cash flow growth to look like under your planning assumptions matching that? Because obviously it's going to be a function of basis differential and the prevailing gas price at that time. So what are your planning assumptions as you see going through 2020 in terms of cash flow growth?

I think as you get out longer term and you see the expansion in the infrastructure that we see as being built and will be built, we don't see Waha continuing to trade at a big discount on down the road, because that will be solved. Part of it starts with the Kinder [Morgan] Gulf Coast Express Pipeline in 2019 and so forth. So I mean historically, Waha has traded at a, call it, $0.10 to $0.20 discount. Over the long haul, you're probably going to be in a $0.35 to $0.50 transportation cost differential at worst. You could – some scenarios actually see Waha become a premium, depending on what happens ultimately with Mexico and so forth and the West Coast. So we think, try to think, longer term with a project like this. We realize it's a little disruptive. If you look at last year when we were putting in a lot of our hedges, we were a lot of the market at Waha. And a lot of the reason why that differential is where it is today. We recognize that. So any time you bring on a world class play that's going to be a little bit disruptive, you've got to go through that time period till you can see through it. But I think as we get out to the timing, when we start to see the cash flows really ramp up here, the beauty of it is we will be solving some of the short-term obstacles that would stand in its way initially.

I appreciate the answer. My follow-up, I'm afraid, is a midstream question also. Because obviously this is a fairly pivotal event if you are able to get something done this year. Your CapEx guide assumes $500 million for midstream. And forgive me if I'm wrong. I think you had indicated you might see the same spend next year also. Are you pretty much done after that in terms of midstream build-out? And if you did manage to find a structure for the midstream, would that capital move off your balance sheet onto another entity, let's say?

Yeah. Well, first of all, next year's spend, we didn't guide to $500 million. We said it would be $500 million this year and $500 million in 2019 and 2020, split evenly amongst those two years. So you're really $500 million, $250 million, $250 million is terms of how we see it. But absolutely, Doug. We envision moving the future CapEx spend into the entity, where it will be able to do its own thing. So that would be the plan. And we're very confident we're going to be able to do something. I mean it...

Really helpful. Thanks a lot, John. Thanks, guys.

There are no further questions at this time. Are there any closing remarks?

No, thanks, Thea. That'll conclude the call. If anybody has any questions, please call myself or Patrick Cassidy. And we'll look forward to speaking with you next quarter. Thanks.