Ladies and gentlemen, thank you for standing by, and welcome to the PAA and PAGP Second Quarter 2017 Results Conference Call. [Operator Instructions] As a reminder, today's conference is being recorded. I'd now like to turn the conference over to your host, Roy Lamoreaux. Please go ahead.

Thank you, Gale. Good afternoon, and welcome to Plains All American Pipeline's Second Quarter 2017 Earnings Conference Call. The slide presentation for today's call can be found within the Investors Relations and News & Events section of our website at plainsallamerican.com. During today's call, we'll provide forward-looking comments on PAA's outlook. Important factors that could cause actual results to differ materially are included in our latest filings with the SEC. Today's presentation will also include references to non-GAAP financial measures, such as adjusted EBITDA. A reconciliation of these non-GAAP financial measures to the most comparable GAAP financial measures can be found within the Investor Relations and Financial Information section of our website. We do not intend to cover PAGP's results separately from PAA's as PAGP's results directly correspond to PAA's performance. Instead, we've included schedules in the appendix to the slide presentation for today's call that contain PAGP-specific information. Please see PAGP's quarterly and annual filings with the SEC for PAGP's consolidated results. Today's call will be chaired by Greg Armstrong, Chairman and CEO; Harry Pefanis, PAA's President; Willie Chiang, Chief Operating Officer of the U.S.; Al Swanson, Chief Financial Officer; and several members of our senior management team are also available for the Q&A portion of today's call. I will now turn the call over to Greg.

Thanks, Roy. Good afternoon. Thank you for joining us on short notice. Today's call will follow a different format than our customary quarterly results call. We will certainly address questions you may have regarding the second quarter or related topics. But to make the best use of our time today, I intend to focus the majority of my comments on addressing modifications we made to our Supply and Logistics segment guidance for the second half of 2017, providing an update on our 2018 preliminary forecast and discussing the steps we are taking to address the disappointments and challenges related to those modifications. With respect to the second quarter performance, earlier today, we reported second quarter adjusted EBITDA of $451 million. As reflected on Slide 3, these results were in line with the guidance provided in May as well as our expectations on a segment basis. We continue to progress our capital projects, which remain on time and on budget. And relative to last quarter, we increased our estimated 2017 expansion capital cost by approximately 5%, which was principally associated with the addition of our new Permian to Cushing takeaway project and related Delaware basin expansion, the details of which were provided in the press release in early July. Although second quarter operating and financial results were in line with guidance, our updated 2017 guidance reflects a $185 million reduction versus the guidance we issued in May. This 8% reduction is predominantly in the margin-based Supply and Logistics segment. There were slight adjustments to our Transportation segment, primarily associated with Permian-related timing issues, modestly lower volumes on our Rockies takeaway asset and the assumed impact of lower oil prices on our pipeline loss allowance barrels. Overall, drilling activity levels and reported well result are in line with to ahead of our…

Thanks, Greg. I would note that we've included our typical earnings update materials in the appendix of today's presentation. [Operator Instructions] Operator, we're now -- Gale, we're now ready to open the call for questions.

[Operator Instructions] Our first question comes from Jeremy Tonet with JPMorgan.

Greg, I just want to touch base on the EBITDA guidance change here as far as S&L is concerned. I was hoping you might be able to dive in a bit more what -- some of the specific drivers might be there. And is this more on the NGL side or the crude oil side? Just any color there would be helpful.

Jeremy, it's both. And I'd probably say in terms of some of the margin and arbitrage opportunities, as I mentioned, in May, we certainly had forecasted what we thought was at the bottom end of what we've been realizing over the last 3 years. And here we are 3 months later, no visibility and -- so we felt like we really needed to make that adjustment. It's a little bit like seeing the frog in the water story, where the heat has been coming up and it's 1 degree here, 1 degree there. When we look back, clearly, we're seeing changes that are somewhat structural, and we're realizing we're not the only ones impacted by it. There's others that are in the community that tried to capture on those arbitrages. We're able to do it with our assets without taking any risks, but we're simply not seeing the opportunity.

Okay. And granted -- just for illustration purposes only, and distribution reductions are a difficult topic. Just curious as far as your methodology in thinking about what the right level is should you look to reduce here. And why not take a larger cut, say go to $1.50 and just take all the pain and really kind of deleverage and move to where you want to be quicker?

Well, I mean, we're -- first off, as you said -- and thank you for acknowledging it's for illustration purposes only. The math, quite simply, is to take DCF as we would normally calculate it, exclude the S&L contributions so that what's left is the DCF associated with the fee-based fully burdened by our interest in maintenance CapEx, et cetera and then determine what's the right level for that. We used the $1.80 as an example simply because that leaves us with 110% coverage, depending upon where we end up on the S&L. If it's within the range of $100 million to $300 million, that would add 7% to 21% incremental coverage. And so if you use the midpoint of that, you'd be at 125% coverage. We've got some asset sales and other things we're doing to reduce debt. And then what we really want to do is make sure that we have visibility for sustaining that and for growth. So when you look at our profile of growing the distribution or growing the fee-based contribution, that should lead to reasonable growth. So trying to find the right balance of what's the right distribution level to protect, quite candidly, people that rely on that distribution and then also come up with the best valuation. And it's a balance with trying to assess that from a coverage level and a visibility for future opportunities for growth. In an ironic world, if we look at our peers that already have achieved that, okay, and we're not there, but if it's $1.80 and it's 110% covered by fee-based distribution and you've got Supply and Logistics adding to that coverage, if you put a 6% yield on $1.80, you could get to a higher unit price than we currently have. I think part of that's because we're not getting credit for -- and today, I couldn't argue hard for it, for the contributions that come from S&L. And so we're trying to find what's the right level to give us the best effective cost of capital, to also honor the fact that people buy our units because they expect a quarterly distribution, and of course, they want as much as they can get. We just need to manage it for the long term. So I think you've asked the right question. Again, we're not here today to say this is the answer. We're here to tell you we think we are going to migrate toward -- or more than think, we plan to migrate toward a fee-based DCF approach to the distribution going forward, which should remove any doubts about the sustainability and should add to it the visibility of growth as we continue to grow what clearly is easier to manage part of the business, which is the fee-based-driven part of it.

And we'll go to the line of Christine Cho of Barclays.

On the S&L guidance for next year, can you provide some color on what you need to see to hit the $100 million low end of the range versus the $300 million? And then data points seem to indicate that we could be tight capacity sometime in the second half of next year, and we haven't seen a large pipeline announcement yet, which I would think would be positive for S&L with respect to supply activities. But you must be seeing something else if you're still calling for margin erosion next year. So curious if it's because production is not going to ramp as fast as we think, there's just too many more expansions we could do or you've contracted out most of the pipeline capacity to third parties, which leaves less for S&L. So any color on what's driving everything would be helpful.

I'm not sure how many questions are embedded in there. We'll give it our best shot, Christine. First off, we haven't -- we don't see something different than I think you described as far as the fundamentals. We expect to see volumes, especially in the Permian Basin, continue to grow throughout all of next year. We do think that the capacity will get tight, and we think there'll be an opportunity for some of the regional basin spreads to come back. What we've been proven wrong for the last year or so is there's other things that happen, and we can't necessarily forecast them. And so by going to this approach of simply giving a range of the $100 million to $300 million for next year and then, excluding from that, determination of what will drive our distribution, the S&L segment, we don't have to spend, quite candidly, 90% of our time talking about 10% of our business. And I'm hoping that as we go forward and clearly understand it here today, but 2, 3, whatever it is, 4 conference calls from now, I hope we spend 90% of the time talking about the fee-based, which is 90% of our business. And so we're hoping to remove that chatter. With respect to the issue of what does it take to get to the $100 million yield, this year, we're showing with -- as I mentioned, in our first quarter conference call, we had some kind of onetime events that were, order of magnitude, about $30 million that we believe we've addressed that should not be recurring. And if you add that back to the $75 million, you're going to come back to right at about the $100 million level, which would say if we have a repeat next year of what we've had this year, in general, realizing there's a lot of moving pieces, that $100 million should hopefully frame the bottom end of that range.

Okay. And then along those lines, I realize this is a very draconian situation, but what sort of things would we have to see for S&L on a full-year basis to be negative? I'm just trying to get a sense of what the downside here could really be and if there is a scenario in which that business to ease into the contribution from the fee-based businesses.

So I think I can address it this way. First of all, the Supply and Logistics business is a moneymaking business. There -- and it facilitates our ability to move barrels on our pipeline. In certain situations, we may choose to take a little bit of a hit on the S&L, but you'd see a corresponding increase in the fee-based side. So if we choose to move barrels on our own systems at tariff levels, but the net realization will be positive for the company but it may end up showing a negative in the S&L. So that's why when I made my comments that we would be largely excluding S&L from the calculation, the math will be what I said. You'll exclude it. The level of coverage that we'll determine is appropriate will be somewhat a fudge of those interaction between the segments. So if there's a loss showing up in S&L because there's a much bigger profit showing up in the pipeline side of it, we'll try to balance that in our discussions going forward, but would not expect -- it's not where we're losing money on the S&L as a company, it's where basically the S&L is taking a bit of a hit to the benefit of the pipeline side.

Or the facilities.

Or the facilities, I'm sorry.

Our next question comes from Shneur Gershuni with UBS.

Just wanted to revisit one of Jeremy's questions with respect to the distribution policy. As you sort of thought about the levels, 1.1 coverage, on the fee-based business, did you sort of take a look at whether you were funding all of your equity component of CapEx on a go-forward basis? Kind of want to think about it in context of are you going to need equity on a go-forward basis? And maybe as part of this discussion, if you can comment on whether -- I realize the board has signed off on it. But has the agency signed off on this as well, too?

Well, in -- let me understand, so that -- the board has basically reviewed the plan that we've outlined for you here today. We -- and they've endorsed it. They're supportive. We have not asked for specific approval, Shneur, to -- with respect to exactly what the math will be and exactly what that coverage level will be finalized. That's my reference to -over the next 60 days, we expect to finalize that. My intent in trying to give, as Jeremy so appropriately pointed out and we appreciate the acknowledgment of that, was that -- for illustration only, was so that we could give an order of magnitude relative neighborhood that we would expect to come out in. That's not to say that if we have a lightning bolt from the sky that tells us here's a much better answer, we wouldn't revisit that. But we're not talking about something that's going to take us to the floor on that distribution. We want to find that balance between what will allow us to have good headroom in coverage on a DCF basis for fee-based and then what will -- and S&L be an incremental additive to that and then get us the -- also the best optimal trading value for the security. So there's a band in there where you move too far one way or the other, it becomes self-defeating, and so we don't want to go outside of that band. So it's an art, not a science, but it's one that we'll be studying over the next 60 days.

And just to clarify in the art that you're painting a picture of, I mean, is the ultimate goal to not be funding anything with equity on a go-forward basis?

No, I wouldn't say that. I mean, if we end up with $1 billion project, we're not going to turn up and say, "Gosh, just cut the distribution." I think we're still able to access the market and -- but I don't think we're going to be hugely relied on. We've been tailing off. As you may recall, our capital program this last year was -- for this year is going to be around $950 million, and it involves some pretty good projects. And then the year before that, it was about $1.6 billion, I believe, and before that, it was around $2 billion. So I think what we're saying is we've got a lot of capacity to fill a lot of pipes, and so we got a lot of cash flow. What -- how we choose to balance that and say we're not coming to mark, I would say this, that if -- if you think of the number as, and let me just pick one, $300 million, if we had to issue $300 million of equity in a given year, we're talking about a minuscule amount of the total company's issuance. And yet it would fund, along with cash flow retained and a reasonable level of borrowing, once we're in our credit metrics, quite a big capital program could easily fund over $1 billion capital program. So I don't think the focus should be totally 100% self-funding. But at the same time, if we don't have better use for it, what a great way to do it.

All right. And as a follow-up question, in your prepared remarks you also mentioned that you were looking at JVs, potentially, asset swaps and listing asset. When you think about the fact that you're looking at the Midland to corporate solution, would you possibly consider partnering with somebody where they would take an equity stake and, let's say, your existing Cactus Pipeline as well also? Or would everything be done on a discrete basis?

I'm not sure I understand what the discrete mean. Yes, I'm sorry. I missed the point there. I think we're open to smart deals. And if there's a smart deal out there, we're willing to do it.

We'll go to the line of Gabriel Moreen with Bank of America.

Just in terms of the illustrative distribution. Clearly, we're halfway through 2017. Can you talk about whether or not the distribution for the next quarter you're looking through to 2018 in this step-up in fee-based cash flows? Or does the 2 next quarters' distribution really have to be based on '17 alone? And I appreciate you're not the board of the directors, but if there's any color on that.

Yes. I think if you look at our second half guidance -- and again I'm going to acknowledge we've had to revise it, but we've tried to get it down to a level where I think our second half guidance, even at the current distribution level, is right at one-to-one. So Gabe, it's -- we're showing a ramp. We've got some projects we're finishing. So we'll look at a combination of the static as well as what we know is happening. So for example, we're finishing Diamond Pipeline. That's scheduled to come on stream in the fourth quarter, I think actually December 1. That's going to kick in cash flows. We've got some step-ups. So we won't do it in a vacuum, but we will be looking forward to 2018 and probably more so for the coverage incremental throughout that. But it's not a huge ramp throughout '18 because a lot of these projects are coming on stream early in the year.

Okay, that's helpful. And then maybe you can talk about -- you talked about strategic acquisitions and still playing offense here. To the extent that there are assets out there, can you talk about the new approach of the assets that are still out there and how they might fit, whether you feel like you might need those assets? And then maybe also talk about that within the context of how ACC has been performing in the second quarter and so far in July.

Yes. As far as acquis, I think what we bring to the table is a lot of synergies and connectivity and the ability to leverage a big system. What we've been struggling with and any -- and everybody has is there's a lot of capital chasing, a lot of acquisitions. So they're going for some big numbers. And in some cases, Gabe, we've missed on the acquisition, but we wanted to own it. And we thought we had more synergies [ our best ], but we just didn't -- going to invest as much capital. I don't know that the ones that we lost, I don't necessarily think they were a bad acquisitions. It's just I don't know that they're going to make as much money as we could have. But again, we saw the risk a little bit differently. As far as ACC, it certainly had, probably in general, like the entire Permian Basin, some of the challenges associated with the delays. But as far as the rig activity level and the overall well performance, we've been pretty pleased, Jeremy. I think it's a fair way to -- so there's -- we certainly wouldn't undo that acquisition, and we're really glad that we got it is probably a better way to say it.

The next question comes from Brian Zarahn with Mizuho.

Greg, in the Permian, could you elaborate a bit on the rising DUC inventory and the frack crew issues, could these be variable that could impact the extent of the volume recovery next year?

Jeremy, dive in.

Yes. Thanks for the question. This is Jeremy Goebel. Yes, it is rising. The completion crews and the activity haven't kept pace with the rig ramp. Consistent across the basin, we're seeing roughly a 5-month lag between when a rig comes online and production. So when you think about the rig count starting to level off at this point, you've got another 5 months before -- at least, before completion activity catches up to where the rigs are. So you're going to see continued build with that gap will start winnowing. So you really won't -- unless commodity prices fall materially and rigs starts to fall, you're not really going to see -- you dig into that DUC inventory until at some point next year. But what you're starting to see is equipment, crews, [ fan ] catching up with the actual rig. And so we will see a ramp, a material ramp in the -- and we're seeing it in our business and you're hearing on calls from the upstream guys, that you're seeing a ramp towards the end of the third quarter, material ramp in the fourth quarter, expect to continue through first quarter of next year. So we think that provides a good support for 2018 activity levels because now you've got an even lower-cost production. So we think that activity will carry you through there. And if you look at flat prices now and as Greg talked about some of the contango arbitrage opportunities going away, that's because producers are hedging next year. That will also provide support for next year. So we think both of those provide support for next year's activity levels.

Yes. And Brian just to elaborate a little bit. When Jeremy says that there's a catch-up on the DUCs, what that means is you quit adding to the DUCs. It won't necessarily bring the inventory down. So effectively, you're 5 months away. If the rig count flattens out before we catch up to one-to-one, a well drilled equals a well completed. So from a standpoint of -- there's appears to be some healthy tension in between the producers and the suppliers that provide the completion services. And what we're hearing is that at -- what's been $45 that's recently risen, nobody wanted to say, "Look, bring an extra 2 or 3 frack crews out there." Because the answer is when you do that, it's going to raise the price. And people tend to raise the price not for the incremental 3 that they bring out, but for all of the crews that they have out there. So yes, I think it's a little bit of a circular issue. I think there's enough resources that if you wanted to start knocking those DUCs down and not only catch up, but decrease them, capitalism works and it'll happen. But it's a little bit of a price tension out there right now.

Appreciate the color in the Permian. Switching to asset sales, what are the next steps in the pending asset sale? And is there a Plan B if that doesn't close?

Yes, Harry, Plan B.

Why don't I go ahead and take that? Greg highlighted we've got about $1.2 billion that we have mentioned before with $900 million of it completed and another $300 million that we expect to close. It's not closed yet, but far in the process, going to regulatory review. He also highlighted we've got $400 million to $600 million additional asset sales that we've committed to. And what I'll tell you is of the $400 million to $600 million, most of those are fairly well advanced. We actually have signed one of them today for roughly $250 million. I won't disclose what it is because we got to respect the process. But the low end of that range, the $400 million, I am very comfortable. We've got conversations that'll take us to that. So I think as far as the Plan B goes or maybe Plan B beyond that, but we are pretty well positioned as far as the $400 million to $600 million. Hopefully that helps.

And beyond.

It sounds like the additional portion -- or the additional asset sale would -- could cover the pending asset sale if it doesn't move forward.

Yes. I guess the other thing we have is we've got some others that we haven't disclosed. We do have a number of other things we're working on. So there's -- we mentioned there's additional asset sales beyond the $400 million to $600 million that we're looking at. So there's clearly some additional discussions around additional volumes leaving the Permian up to Cushing and perhaps some others on the West Coast.

Pending sale, I mean, if for some reason it doesn't go through, there is a Plan B.

Yes.

Our next question comes from Vikram Bagri with Citi.

Just wanted to spend more time on S&L segment. You mentioned that the lack of visibility margins is mostly leading to a cut in S&L guidance. My understanding was most of this year's EBITDA was somehow protected using summer/winter spreads on propane, crude basis differentials, and a large portion of that EBITDA, substantially all of that, was coming in fourth quarter. So were there 1 or 2 items that you can point us to, which resulted in change -- in such a significant change in guidance in S&L segment?

There are a handful of things that are -- even -- look at the back of the curve, I mean, the contango margins have narrowed, lots of crude grade differentials have narrowed. We always -- when you look at sort of the NGL business, we have a base level of business that's locked in, but we optimize within those strategies with respect to location differentials, optimizing transportation. Now a lot of those things that we've historically seen have narrowed. And as Greg mentioned, we just don't have a clear visibility to be able to drive those in the third or fourth quarter of this year.

Vikram, I would add to it. It's -- and we're suffering our own fate that we kind of put out there is -- when you're running well above one-to-one coverage, and we've seen some of these opportunities not be visible in one month and certainly become very visible the next month. And if you look at the slide that I think we have that shows the quarterly performance of -- on Slide 7 in the upper right, you can see that there's a fair amount of movement in the Supply and Logistics on a quarterly basis. What -- when we're running less than one-to-one and we're also trying to manage our balance sheet for our credit metrics, what we did not want to do is close our eyes and hope that what has happened in the past continues to happen in the future. These opportunities pop up. I would tell you that there's certainly an opportunity for additional opportunities to come up in 2017, later in the year, and 2018 that could cause us to look back and say, "Gosh, why did you lower the guidance?" But again, when we're sitting here where we are, we're looking -- we said without the visibility, the most transparent straightforward thing we can do is recognize the market has changed. The differentials, as we talked about a few years ago, went upside down, when you had barrels that were priced to where they shouldn't go to any market besides Midland and yet because all the takeaway capacity was at prices greater than what that differential was. So there's things we can't control. A lot of capitals come into the system, the distortion of the MDCs that makes it more difficult for us to be patient and say, "Even though it's not visible, it's probably going to happen." As I mentioned earlier, just to give -- put it in perspective, the last 3 years, what we forecasted for 2017 was at the lowest end of the last 3 years and probably about 40% below the average of that year. So we thought we were fairly safe coming into it. And now where -- that was at the end of April. So now we're at the beginning of August with only a little bit left. And these transactions require also realization within the calendar year. So if we do a deal in November, December, it's -- likely, it's going to push into 2018 as well. So you call it abundance of caution that was [ light ] or whatever, but I don't think necessarily we're saying that there are not going to be opportunities. We're just saying we don't have visibility for them.

Okay. I wanted to make sure that some of the S&L EBITDA did not moved to a fixed-fee segment. Though you did not change the guidance for fixed-fee segments, there is no EBITDA moving from S&L to those segments and probably there's upward bias in your numbers for fixed-fee segment. Is that...

There's -- Vikram, there is some. I mentioned -- somebody mentioned earlier, said how could you have a negative S&L. The answer is if we were to go out and basically augment the fee-based movement at the risk of S&L, but there's not a material amount that's moved there at all.

Okay. And just last one for me. I understand that the line fill through -- in the S&L segment. Can you share what the hard assets are in this segment and what the book value of those assets is?

There -- in addition to the linefill, there are some pipeline segments, some of the trucks and barges and rail cars types of assets are also in S&L. I don't have the numbers in front of me, but I think we've got north of -- gosh, maybe 10 million barrels that's in line fill, that's in the S&L segment, probably closer to 12 million or 13 million barrels. Don't hold me to that, Vikram, but our 10-Q will be up. But I think it's upwards of 10 million to 13 million barrels in that segment. So there's -- it's not without assets. And in fact, it's just liquid cash.

Yes, linefill and a lot of the Logistics asset Harry mentioned that were part of the development asset in that segment.

Our next question comes from Ethan Bellamy with Baird.

Greg, with all this capital flooding in and the competition, what assurances do we have that we're not going to see attrition in the fee-based business? What kind of weird average contract life do you have, hence the MVC scenario? I'm just curious if these S&L pressures are going to end up impacting the fee-based side as well.

Well, I think, first off, I'd say the -- I will say the S&L is usually going to affect it. I would say that the capital has already affected it, so what you've seen is lesser growth than we would have anticipated. And quite candidly, lesser growth both in terms of volumes because of head-to-head volume competition, Ethan, but also, I mean, just pressure on tariffs. And so for example, without giving you the names of the pipelines, we've built some 3 or 4 years ago, where the tariff that we were able to get and secure a 10-year contract on would generate a well into double-digit rate of return, north of 20%, and in fact, was generating that. Whereas new build opportunities, it's hammer and tong out there, and you're going to get probably double digits but low double digits, and you're not going necessarily get it locked in up front. And so I'd say it has impacted it. As far as, if your question is could we see a deterioration from the growth that we anticipate, we've got demand full of pipelines. Diamond is clearly a demand pull with a long-term MVC on that.

Red River's demand pull.

Red River's demand pull, Caddo was demand pull. In addition, we've got -- we had an open season -- or not open -- well, we did have an open season as of -- for 50% of BridgeTex, and we had good interest in the existing capacity because -- especially out of the Permian, because as one of the caller's -- questioners already mentioned [indiscernible], there hasn't been another project announcement. And right now, if you look out, the balances are going to come in our favor. So I can't tell you what happens 3 or 4 years out. I will tell you this, that we are basically back to the new project that we announced from the Permian to Cushing is backed by a long-term contract. We've tried to set our rates, and this is going to sound awkward, at a level that when the contract expires, we won't be the highest tariff out there, so we're probably giving up some ground to have certainty that our pipes will continue to be the most relevant versus those that may have priced themselves up at the upper ends of what can I extract. And then that's a 5-year contract. At the end of 5 years, it could be a rude awakening. So our tariff to Cushing is very competitive. Our tariff, that we've set through our open season, on BridgeTex is very competitive. Other things on Cactus. And as we try to look at building Cactus II, we're taking the same approach. So I think we're talking about 2 different issues in the sense of Supply and Logistics versus the pipeline. I do think it's the same issue, which I think is what you're driving at, just a lot of capital coming in is going to change the landscape, but that's already been happening for the last 3 or 4 years on the pipeline side.

Okay. So can you quantify for us what portion of the fee-based business, let's say, 2018 is open to recontracting when it's not locked into the fee that you know extends beyond the year?

Yes. Well, I can try to address it in pieces. So like on Basin, I mean, which was built 50 years ago, there's very few actual contracts on it, but it's got an $0.80 tariff. And that's a pretty attractive tariff to get to the best market in the world in Cushing. I will -- maybe I'd address it this way. We've gone through and looked at all of our existing contracts, if this is your question, and said, "What's the amount of contracts that we have expiring in the next 1, 2, 3 and 4 years, and how does that relate to our current and projected tariff revenue from that?" And Al, correct me if I'm wrong, I think the amount that we had expiring, Ethan, that would have had exposure to what we thought was above-market rate, which is I think the issue you're trying to get to, was less than 1% of -- on those 4 years.

So all 4 years through, say, 2021?

The average was right at -- less than 1%. I think it was like 0.75% in '18 and then it was -- which is built into, by the way, our preliminary guidance. And then in '19, I think it was like 1.2%, and then in '20, it was 0.8%. So small numbers, in other words.

Okay. And so can you give us a weighted average contract life?

Well, again, realizing I don't -- relative to capacity -- I mean, I've got very limited on Basin, on -- it's almost like an asset by asset. So I think on Cactus, we've got 7, 8 years left on it. BridgeTex has probably got 7 or 8 years left on it. So the step up in the Rockies is probably closer to 4 to 5 years. But again -- so I can't give you the weighted average -- if I gave it you, it probably wouldn't be worth anything because of the apples and oranges comparison. I think the most relevant is at least for the next 3 or 4 years, there's not a big -- there's no cliff coming at us on anything.

Most of the projects were sanctioned 2, 3 years ago in lives of 5 to 10 years. Maybe that will help.

Eagle Ford, I think we got like 6 or 7 years left on those.

I mean, you take pipelines like our Salt Lake City pipeline. There's not many contracts on it, but there's refiners, and that's the only line that takes through into Salt Lake City and LA basin we've got the only 2 lines from -- into the LA basin from...

San Joaquin Valley.

San Joaquin Valley. So those pipes, they don't have contracts, but most of these, they're just shipper history basins and they're really demand pull comp -- or demand pull.

Our next question comes from Ganesh Jois with Goldman Sachs.

So Greg, I think in your comments, you alluded to balancing the requirements [indiscernible] reestablishing your dividend policy. I guess a few observations. I think, firstly, the assumptions [indiscernible] the illustration you've given, firstly, that S&L doesn't go negative, and this quarter, it did go negative. Second, there's no downside to your transport and facilities guidance, which again, it's not very clear that there isn't any downside to that. My question is wouldn't you be better off resetting your distributions to a level that takes all of these risks off the table, knowing fully well that there are several companies out there with essentially higher coverage ratios than what you are modeling on, on a fee-based business?

Yes, just a couple of clarifications, Ganesh. I mean, on the negative coverage, negative S&L, we really talked about it on an annual basis because, clearly, there's a seasonality where we're incurring costs but we don't have any sales for some of the NGL transactions that are hedged into the fourth quarter. So we are looking at it on an annualized basis, not on a in the Street quarter basis. With respect to -- certainly, I would -- maybe I misspoke. If somewhere in there, you got the impression that we said there was no downside to our fee-based business, we're not trying to say that. We said we've got good reason to believe that we've got a solid case for it to be in the range that it is. And then part of the reason that you have coverage from that -- distribution coverage is for what you don't know and what comes at you. So we're just trying to establish that, and if we do that with the fee-based, which is the much less volatile and lower risk, and it's also the one that happens to be growing, as mentioned, it's doubled in the last 4, 5 years and we're looking for another 15% growth there. So we're going to take all that into account. I mean, I don't doubt that before the next 60 days are up, we will have as many people as we talked to, we'll probably get that many opinions on what we should do. And we'll listen to, patiently, as much as we can, everybody's input on that. But at the end of the day, we all got to make a judgment and -- about what we see, and as you might expect, we probably have more data than most people outside looking in. That doesn't mean we have the wisdom perhaps of some of them. But we'll take all into account. We'll make the best call we think is for the long-term. I would reinforce that, again, management, the board and its affiliates represent 25% of the total equity capitalization. So we're going to balance all that to try for the long term to do the right thing.

Got it. And maybe a follow-up question. You've talked a lot about asset claims and you've already executed on quite a few. Kind of as an outsider looking in, I wonder about the wisdom of claiming assets when really, what claims stands for is very integrated and basically hard to replicate asset base. And for you sell off pieces of that asset base maybe not very valuable today but might become valuable down the road, as an outsider looking in, it's very hard for me to understand that given that we really don't know what the future holds and what assets are valuable at what point in time. So again, it goes back to the distribution policy in the context of how you're trying to rightsize...

I appreciate your observation. Again, in some of these cases, the assets that we've sold, we've looked at it and said, "There can't be more upside to it. It's just more valuable to somebody else." There's a heck of a lot of capital out there that's trying to form around fee-based activities, and so there's more downside than there is upside. We're certainly not giving any assets away. We're getting attractive multiples, and in some cases, were getting attractive multiples at full utilization. So again, your points are well taken. Philosophically, I can't disagree on what you said. I think factually, there's sometimes a difference between what we know and what the public sees.

And some of the sales are actually sales with joint venture partners where they're bringing opportunities to the table. So we're expanding the opportunity set by allowing another party to participate in the asset.

And our next question comes from [ Matt Anely ] with JPMorgan.

Maybe a question for Al. Greg, I'd like to hear your thoughts on it also. But just in terms of the rating agencies, obviously, nobody's really talked about that yet. Have you guys run this plan by them? Certainly, Moody's was focused on operational issues, less so than absolute leverage metrics, which obviously, you can address with equity or asset sales, but EBITDA is clearly a focus for them. So can you talk about how you've had conversations with them and how you guys addressed that?

Yes, Matt, this is Al. I'll take a shot at it. Clearly, we previewed the earnings and the guidance revision with agencies. Obviously, second quarter results in line with guidance revision and, clearly, is viewed as a negative development by the agencies. Commensurate with that, we're offsetting that to a degree. We believe they'll look at what we're discussing here on distribution policy, prospectively, as a credit positive as well as the incremental asset sales, how we're managing our short-term debt, et cetera, and some of the other tools we have to try to meet the leverage ratios that we have out there. So they didn't come and "sign off" on our approach or anything like that. As you know, they don't do that. But they did understand it, listen to it and see the negative side of the guidance revision but the positive side of some of the steps we're taking.

Okay. And then just in terms of the balance between potential equity use on a go-forward basis, obviously, you're planning to delever to get to those metrics. Given the asset sales, it's hard for us to see through kind of how much cash flow you're giving up in those asset sales granted you're getting decent multiples for them. So how should we think about that on a go-forward basis in terms of your thought process around the distribution...

Yes. I mean, clearly, if you sell an asset at -- I'll just make up a multiple, around [ 1 10 ] multiple, part of those proceeds are not deleveraging and part are. If you're looking at a 5 multiple, half of the proceeds in effect improve your leverage. Kind of back to Ganesh's question, we feel like what we've sold to date and what we're looking to sell, we've gotten good values. There is a lot of capital out looking for assets. We've been able to sell them to strategic folks that actually can bring synergies to it. So we do think that those have and will provide deleveraging aspects to it. Clearly, when we look at it, as Greg alluded to earlier, our CapEx program is reduced as we look ahead from what it's been trending. Asset sales for the balance of this year will basically fund the majority of our remaining capital. Clearly, a distribution policy change will help coverage, help retain more cash flow. As well as we have other alternatives, whether it's looking at our existing preferred and taking a look at if we can pick that dividend a little longer, raise incremental preferred, raise common equity, so there's a number of other things that we're looking at in concert with this as well. Thoroughly reestablishing coverage makes it, as we're talking about, as Greg mentioned, within coming out with a defined policy in the next 60 days, increases our ability to use, whether it's common equity or preferred equity, as well.

This question comes from Harry Mateer with Barclays.

So Greg and Al, you've talked about how you're changing the way you're thinking about DCF and coverage by excluding Supply and Logistics from the basic calc going forward. And I guess from a bondholder's standpoint, are you willing to do the same thing for targeted leverage metrics and perhaps exclude budgeted S&L EBITDA from your planned leverage metrics?

I can't tell you that we've gotten that far yet to be able to land on it. I think we're -- we probably have -- there's many different people that have opinions about what's the right leverage metric. It's a qualitative issue, and I think you're introducing the issue as what a leverage metric that was 100% backed by fee-based be the same exact metric as something that was backed by something that was 80% fee-based and 20% S&L. And so I would acknowledge that there's a difference. We haven't articulated that publicly; and privately, we're still trying to figure out what the best way to get to the right level that we're comfortable with. And then we got to see how that conforms with what the rating agencies view it. As you know, they turn around and make incremental adjustments that are not related to our GAAP financials or even our non-GAAP financials that were out there. So I think the commitment is that we are very much committed to maintaining an investment-grade structure. It's an important part as we see going forward in managing our cost of capital. We think ultimately, the steps we're taking on the distribution policy will generate incremental cash flows that will be available to reduce leverage. I mean, we've -- even though it's been a challenging couple, 3 years since we entered into late '14, our actual debt level has gone down. I think in 2016 and 2017, projected to go down, and we're expected to go down in 2017, so -- excuse me, '18. So that, combined with the rise in -- even though we're reducing S&L, we're still showing year-over-year growth in aggregate EBITDA. And I think this year, we're at $2 billion of EBITDA from our fee-based. The next year, at 15%, we're upwards of 2.3%. So the combination of decreasing debt and increasing EBITDA is a pretty proven formula for getting to attractive leverage metrics.

Okay. And then just one additional one on the agencies. But how critical do you guys view it to maintain IG ratings at all 3 of them?

Our view is we want to retain IG ratings at all 3. Clearly, we don't control all aspects of that, but our intent is to do what we can to do that within reason. Clearly, we added a third one that kind of provides some assurance that if we did have a misstep and something happened, we would still be in the indexes, but our hope and our desire is to retain all 3.

And we'll go to Danilo Juvane with BMO Capital.

Does the fact that you're maybe, perhaps a little more constrained balance sheet-wise here, does this limit your ability to either grow organically, secure incremental organic growth projects for even M&A over the next year or so?

I think the answer to that is no. If we choose -- there is many different ways right now to finance projects if you've got a good, attractive project just because of the amount of capital that's out there. And so that can be a combination of either strategic partners, financial partners or simply accessing a different type of capital. I think there's different ways to get there. So what really drives that decision on the growth is it needs to be a good strategically -- or something that is not just spending capital and trying hard for a return. It's something that fits within our system. And typically, that comes with a lot of synergies that we can bring to the table that allows us to clear that cost of capital with a reasonable margin. That same attribute is the very thing that would attract a lot of external capital on a joint venture basis or strategic partner or with another industry partner that would want to come along with us there. Long way of saying the answer is no, we don't see a constraint in this.

Okay. As a follow-up question, Greg, you mentioned in your prepared remarks a lot of structural changes to the S&L business. Within the 60-day timeline that you will sort of evaluate your distribution policy, I was curious to know if you ought to be evaluating what the sort of rebase rate for S&L should be given these special changes that you guys are seeing in the industry?

We're going to try. I mean I think one of the things that we were trying to be as forthright as we could at Analyst Day or Investor Day in May, and we got asked the question several times, what -- remind us what the baseline was. And for those on the call that don't recall, we used to run expected baseline of around $500 million plus or minus. And at times, we would run as high as $800 million to 900 million, and we were quick to caution people that we didn't necessarily consider that excess of the baseline to be either recurring or predictable or some combination of that. I think back then, we were called sandbaggers. Today, we may be called something a little bit different. But obviously, there's been a change because of this light amount of capital structural issues that have been changed. And so at the Investor Day, we basically said we're not prepared to marry up to a baseline and because it's still changing. And here we are 3 months later and it's a different perspective. What we're seeing out there is we're not the only one that's feeling some of this pain, and I mentioned earlier that we've got some MLPs, and we're pretty transparent about how we -- where we capture these opportunities in there. And Supply and Logistics, I think in other entities, they group their segments and their business units differently. And that's fine, there's nothing wrong with that. It's just not as evident when there's pressure on that. But we're certainly seeing explanations that would suggest margin compression is not limited to PAA. The other thing I would just say is some of the players that do both not only physical but notional activities in terms of the…

And we'll go to Becca Followill with U.S. Capital Advisors.

First question is on Page 8 of your slide deck. Item 4, you talked about reducing short-term inventory and/or changing the financing method. What does changing the financing method mean?

Either using equity or preferred capital against it. Having intermediation type of -- where basically steps in between it, would be the 2 that would come on the mind.

And what does intermediation do to your margin?

It would take a little bit out of the margins.

Can you quantify in some way?

If 2% cost, call it, goes to 4% cost.

Okay. Second question on S&L. If you're going to go to a fee-based distribution policy, why not just zero out S&L guidance?

Interesting thought. We're -- we got 60 days, Becca, to kind of -- that we put ourselves on the clock for to try and figure out how is the best to communicate with the public. And so far, we haven't got to the zero out. We certainly got to defining the wide range so that we're not bearing to a particular number and somebody's putting that in their model and saying this is what management told us. So again, we're going to capture all the opportunities that are there. We're not going to quit trying. But again, it's been a challenge, and we're trying to make the right mid-course adjustments on the distribution so that, that's not an issue in the sustainability of the distribution. And again, it's painful on the transition. It's painful overall, but especially painful on the transition. So I want to add that to my list of opinions that we'll take into consideration there. I'm not sure exactly what that will do for us, but I kind of like the concept.

Yes. By zero out, I mean, not give up, but just expectations so if anything becomes gravy.

No, I heard you.

Okay. And the last question is the Q3 guidance that the -- would come in EBITDA, 21% to 22% of the $2.1 billion would imply Q3 at about $460 million. But then Q4, up $670 million, which is a pretty big hockey stick, especially it looks like Q3 is flat with Q2. Just given the kind of missteps in guidance, can you just give us some more color around what causes that hockey such a big hockey stick?

Yes. What I would say is that we're modeling transportation segment sequential growth, third quarter over second quarter, fourth quarter over third quarter. And part of it, what you heard is fairly back end-loaded volume ramp, as Jeremy mentioned. So actually a pretty decent amount of growth coming from that segment. Clearly, we've lowered our expectations for the S&L segment for the fourth quarter, but it is seasonal. We still expect to have a much better quarter than what we're modeling for the mid-quarters 2 and 3, where you're actually modeling slight net losses.

And we'll go to Jean Ann Salisbury with Bernstein.

Just one quick one for me. How much variability per year in distribution, if any, are you comfortable with? For example, if transport and facilities is down 5% in 2019, would you let it move around? Or just you would plan and leave enough coverage that you shouldn't have to?

The latter.

We'll go to Matthew Phillips with Guggenheim.

A follow-up on that last question. I mean, do you expect to provide annual visibility on a variable distribution? Or do you expect it to be something more set on a quarterly basis?

First off, we're not looking for a variable distribution at all. What we're looking to do, Matt, is set it at a level that we know we can sustain with hopefully visible growth in there. And again, because we're focusing on just the fee-based, and that's been more -- again, when I got contracts, I got projects, it's easier to forecast that growth. So we're not sure yet whether we'll continue to give -- whether we'd do an annual increase in distribution if the numbers made sense or quarterly, but what we do know is we want to get it right on the front end. And as Jean Ann mentioned, it's to do it in a way where we've got enough coverage that if for some reason something goes bump in the night, it's not being reduced. But hopefully, one that leaves plenty of room for growth as we grow the fee-based part of the business. And we've got, I think at our Investor Day, we went through many of the assets on the facilities and the pipeline that showed that we had significant growth just filling the pipes, and so we still have all that. This chatter on the S&L is just taken away from that focus. So no variable distribution, steady growth when we start growing it, and hopefully, the sooner as we have the distribution stabilized relative to the fee-based part of it.

Okay, so you want to set it up a level low enough to where you're -- on a go-forward basis, it's only based on assets that, in theory, are only going up in terms of volumes in cash flow?

Yes, I never want to have this discussion again.

Yes. And then as I look to '18 CapEx, on the chart you've given, I mean, I'm looking at around, I don't know, $400 million or so in projects that spill over into $400 million to $500 million into '18. I mean, what sort of level do you expect in '18 compared to what you have in '17?

We haven't provided a specific number for '18, but as you know, a number of projects will spill over into '18. It will be, as of now, based on approved projects, it will be lower than what we have for 2017.

Yes. I mean, it seems like it's mostly the Permian and Fort Sask projects. I mean, is that -- does that sound about right?

That's right. And it includes the most recent Permian takeaway in there. Obviously, Matt, we're looking on some other projects that could add to that, but again, those are going to be really happy days if we get those big projects lined up. And that will raise opportunities for different types of ways to finance it as well. Is anything relating to Cactus II in these figures?

No.

Shneur Gershuni with UBS.

Surprisingly, I have more questions. A couple of crossing the Ts, dotting the Is. In your illustration, what was the unit count that you were assuming for the $1.80?

We have round numbers, and one of the guys can correct me. I think about 730 million, [ the short is ] 730 million on the common. And then the preferred, which is currently being picked and has a [ 2 10 ] distribution level, that's roughly about 69 million to 70 million.

Okay, perfect. And then 2 little quick follow-up. With the basin open season being a success, shouldn't the guidance for fee-based have gone up for 2018?

The line doesn't go into service until -- yes, until...

Early '19.

Oh, early '19. Okay, that makes sense. And then finally, in one of your responses to the questions about potential losses or costs within the S&L business, you had mentioned you have some rolling stock and so forth. When I think about the 5-year leases that were entered into and so forth over the past bunch of years, I would have assumed that there aren't that many that you've entered into since the oil prices collapsed. When do we hit a point where some of that rolling stock goes back off lease, and so forth, and we start to see some cost reduction in that respect?

Are you talking about railcars and...

Yes, exactly, railcars.

Well, it's coming. I don't have the numbers right here in front of me, but obviously, that would -- it's going to -- to the extent that their idle cars is going to add to the S&L margin, if you will, just because of the removal of the cost. We can circle back around to you there. I don't know that. I just don't have it right here.

We'll go to Patrick Wang with Baird.

Yes. Just really quick. Going back to the Alpha Crude Connector business, you touched on it a bit earlier. Is it fair to assume that 2Q results now reflect the full initial asset ramp? Meaning going forward, can we expect that the ATC system should now grow in line with overall Delaware or regional activity there?

Directionally, I don't think that's a bad statement. I mean, part of the challenges when we go to these pads, if you have 2 6-well pads come on in a quarter, the answer to your question is no, because it's going to jack it up in a hurry. And that's what we're seeing right now. We're seeing some delays in certain areas where what we thought was going to come on August 1st is now going to come on November 1. And if it comes on the same time that another pad that was supposed to run [ for a little bit ] stays there, you could have some soft tooth kind of reaction on some of these systems. The good news for our producers and for us is because of our integrated approach, we could handle those volumes, but it does take a lot of moving pieces. Because if you have 6 wells come on at 1,000 barrels a day-plus each at the same time, it can -- and normally, you would size your lines different if it was just a single well completion. So I think you can get some erratic, but generally upward bias, to that. On the downside, there is some prevailing practice now showing up in certain areas, especially in Oklahoma, that when somebody, your neighbor, fracks their well, you shut yours in and because you don't want to potentially have a pressure sink and pull it across. We haven't seen a ton of that, I don't think yet in the Permian, but it's possible we could start seeing that in the future. So you could see some [ erratic ] stuff from time to time. It doesn't mean the reserves aren't there, it just means you're trying to do the most prudent thing operationally.

Okay. And then do you believe there's enough excess capacity at this point to handle those, call it, those spike volumes, let's say, in the fall or later in your time frame?

We've got a SWAT team that their sole job is to make sure we don't ever say no to a producer.

Sunil Sibal with Seaport Global Securities.

Just one quick one for me. When you look at the various possibilities that you're kind of looking at over the next 60 days, and then when you look at your target leverage of 3.5 to 4x on the long-term debt, I was curious to know what kind of time will be that you're looking to hit at to get to that leverage?

To get to our long-term debt-to-EBITDA metric?

Yes.

Yes. We won't and we didn't expect to probably be at that level at year-end 2017, but we expect it to be at that level as we progress throughout 2018.

And our final question comes from Selman Akyol with Stifel.

I actually have 2 quick ones. First of all, in terms of the preferred next year, did I hear you think maybe you're extending the time period for pick or trying to go back and open that up for pick?

That is probably something that will be discussed, yes.

Clearly, the holders would have to agree to that contractually, and we don't control that by ourselves.

And Selman, by the way, the pick extends through the first quarter, so we'd be talking about the last 3 quarters of...

Yes. Through the February 18 payment, and then it turns to cash pay with the May 1.

All right. And then just kind of given the level of competition that's out there, and I'm trying to just sort of balance strategic versus dilutive. What kind of returns now do you have to have as a minimum in terms of going out and either doing a project or doing something a little strategic. Just when do the returns get to be to the point where it just isn't worth it for you guys even if it is strategic?

I mean, we're generally looking at things that are 400 basis points-plus above cost of capital, sometimes higher or lower, depending on what the risk is. And the cost of capital can move around a little bit. But generally, you're just talking about low double digits. Part of the issue, Selman, if I'm interpreting what you're trying to ask correctly is, what do you know at the day that you sanction it and that you think you're going to get at a minimum. And what do you think you're going to get 2 years out if you're call on -- our call on production volumes growth is right? So for example, you may turn around and accept a minimum rate of return that says if I never get another commitment, I'm going to get my cost of capital plus 100 basis points. But if we are able to fill 3/4 of the pipe up because we're the best connectivity, and we've got competitive levels on the marginal tariffs that, that number goes into a high-teens kind of return. That's a pretty attractive proposition for a strategic investment. So part of it is, what's your view and what's your level of commitment? That's why we spend so much time on the fundamentals, because in this day and age, it's harder to get people to commit based upon what they think they're going to produce 4, 5 years from now. And so they're willing to make some commitment, but not necessarily underwrite the entire pipe the way it may have been happening 6, 7 years ago.

We have no further questions.

Thank you very much. Appreciate those that dialed in. Obviously, it was a late call. We wanted to accelerate the delivery of the information as fast as we could. We appreciate you spending the time, and we appreciate your support. Good night.

Ladies and gentlemen, that does conclude your conference for today. Thank you for your participation and for using AT&T teleconference service. You may now disconnect.