Good day, everyone, and welcome to the second quarter 2013 ONEOK and ONEOK Partners Earnings Conference Call. Today's conference is being recorded. At this time, I would like to turn things over to Mr. Andrew Ziola. Please go ahead, sir.

Thank you, and good morning, everyone, and welcome to ONEOK and ONEOK Partners Second Quarter 2013 Earnings Conference Call. A reminder that statements made during this call that might include ONEOK or ONEOK Partners' expectations or predictions should be considered forward-looking statements and are covered by the Safe Harbor provision of the Securities Acts of 1933 and 1934. Actual results could differ materially from those projected in any forward-looking statements. For a discussion of factors that could cause actual results to differ, please refer to our SEC filings. Our first speaker is John Gibson, Chairman and CEO of ONEOK and ONEOK Partners. John?

Thank you, Andrew. Good morning, and thanks for joining us today. As always, we appreciate your continued interest and investment in ONEOK and ONEOK Partners. Joining me on the conference call today are Derek Reiners, our Chief Financial Officer; and Terry Spencer, our President. Also on the call and available to answer questions are Pierce Norton, our Executive Vice President of Commercial; and Rob Martinovich, our Executive Vice President of Operations. On this morning's call, we will review our second quarter 2013 financial results, update our point of view on the magnitude and duration of ethane rejection, and explain why it should have no impact on our 2013 or our 3-year earnings expectations, review our progress on our growth projects, and to avoid any further confusion, we will reiterate the ONEOK dividend growth we expect once the -- once we separate our natural gas distribution business. Let's start with our second quarter performance. ONEOK's second quarter performance was solid after adjusting for the noncash charge related to the wind down of our energy services segment. We had a robust volume growth in ONEOK Partners with volume increases in natural gas gathered and processed, and natural gas liquids gathered, all as a result of the growth projects we completed. In ONEOK Partners, our NGL exchange services margins continue to increase, compared with the same period in 2012, as we execute on our strategy to convert capacity held for optimization activities to fee-based contracts. As a result, and as expected, our optimization margins declined as a result of this reduced capacity. And the expected tighter NGL price differentials were experienced between Mont Belvieu and Conway. As anticipated, ethane rejection had some impact on our second quarter results. However, it was less than $15 million. Our natural gas distribution segment performed exceptionally well during the quarter, benefiting from new rates in all 3 of the states it serves and from higher sales and transportation volumes related to the colder-than-normal weather. As Terry will discuss in a moment, we are revising our outlook for ethane rejection during 2014 and 2015, based on recent developments in the NGL market. We expect that current levels of ethane rejection, which are less than we originally forecasted in February, will continue into 2014 and into 2015, but at lower levels. We remain confident that we will meet our 2013 guidance expectations and our 3-year earnings growth expectations, averaging an annual 15% to 20% EBITDA increase during that period, even with ethane rejection continuing through 2014 and 2015, due primarily to additional volumes now expected that were not originally included in our 3-year plan. Derek will now review ONEOK's financial highlights, followed by Terry, who will review ONEOK's operating performance.

Thanks, John, and good morning. ONEOK's second quarter earnings were approximately $1 million or $0.00 per diluted share, including the noncash after-tax charge related to the accelerated wind down of the energy services segment. ONEOK's net income in the second quarter would have been approximately $72 million or $0.34 per diluted share, excluding the noncash charge, compared with $61 million or $0.29 per diluted share for the same period last year. Improved second quarter results reflect volume growth as a result of completed growth projects at ONEOK Partners in the ONEOK Partners' segment. The natural gas distribution segment posted higher results due to increased rates in Oklahoma, Kansas and Texas, and colder weather in the service area compared with the same period last year. As we announced in June, ONEOK will discontinue the energy services segment through an accelerated wind down process. Energy services released a significant portion of its nonaffiliated third party natural gas transportation and storage contracts to third parties in June, effective July 1, 2013, and recorded $114 million pretax noncash charge. We also expect about $40 million of additional charges later this year or in the first quarter of 2014, related to the remaining capacity that we intend to release. We expect future cash expenditures associated with the released transportation and storage capacity from the wind down to total approximately $100 million on an after-tax basis, with approximately $15 million paid in 2013, $35 million in 2014, $25 million in 2015, and $25 million over the period of 2016 through 2023. As John mentioned, ONEOK also affirmed its 2013 net income guidance range of $235 million to $285 million, updated for the energy services current and expected future after-tax charges related to the wind down provided in June. ONEOK also affirmed its 3-year expectations of net…

Thank you, Derek, and good morning. Let's start with our natural gas distribution segment. As you read in the news release, second quarter 2013 earnings were higher, primarily reflecting higher rates in Oklahoma, Kansas and Texas, and colder-than-normal weather in the second quarter 2013, compared with warmer-than-normal weather during the same period last year. Operating costs were slightly higher compared with the second quarter 2012, primarily because of higher pension expenses and property taxes, which were mostly offset by lower share-based compensation expenses. The midpoint for the natural gas distribution segment's operating income guidance for 2013 remains at $227 million. As Derek discussed, the energy services segment incurred a noncash charge related to the released capacity in the second quarter. Excluding that charge, net margin decreased by approximately $14 million in the energy services segment compared with the same period last year, primarily due to lower realized seasonal storage differentials and marketing margins, net of hedging activities. 2013 guidance for the energy services segment is a loss of $210 million, that reflects the noncash charges related to the accelerated wind down, unchanged from what we said in June. John, that concludes my remarks for ONEOK.

Thank you, Terry. Now Derek will review ONEOK Partners' financial performance, and then Terry will come back and review the partnership's operating performance, growth projects and discuss in a little more detail the reasoning behind our revised ethane rejection outlook.

Thanks, John. In the second quarter, ONEOK Partners' net income was $202 million or $0.62 per unit compared with $207 million or $0.69 per unit in the second quarter of 2012. Distributable cash flow was $252 million in the quarter compared with $240 million in the second quarter of 2012, resulting in a coverage ratio of 1.17x for the second quarter 2013. By comparison, our coverage ratio was 1.29x in the second quarter of 2012. Distributable cash flow was $445 million for the first 6 months of 2013, providing a 0.99x coverage compared with $519 million for the same period last year, providing coverage of 1.50x. Our long-term annual coverage ratio target still remains at 1.05x to 1.15x. However, as we've discussed in previous conference calls, we expect our full year 2013 to be slightly above 1x coverage. ONEOK Partners affirmed its 2013 net income guidance range of $790 million to $870 million announced in February of 2013. The partnership's DCF is expected to remain in the range of $910 million to $1 billion. Operating income in the natural gas liquids segment is expected to increase, while operating income in the natural gas gathering and processing segment is expected to decrease from previous earnings guidance. Terry will give the specifics in a moment. We affirmed EBITDA to increase by an average of 15% to 20% annually over a 3-year period comparing 2012 results with 2015. ONEOK Partners also affirmed that it still estimates an average annual distribution increase of 8% to 12% between 2012 and 2015, subject to ONEOK Partners' Board approval. Capital expenditures for 2013 are expected to be approximately $2.36 billion, comprised of $2.24 billion in growth capital and $120 million in maintenance capital. This update reflects the timing of capital dollars and the movement of some capital…

Thanks, Derek. The natural gas gathering and processing segment's second quarter operating income was almost 20% higher, due primarily to higher natural gas volumes gathered and processed, offset partially by lower realized NGL prices and higher compression and operating costs due to the growth projects we've placed in service over the last year. Natural gas volumes gathered and processed continue to grow, driven by increased well connections in the Williston Basin and Western Oklahoma. For the second quarter, natural gas volumes gathered increased 23%, and natural gas volumes processed increased 28%, compared with the same period last year, driven by the new Garden Creek and Stateline I and II natural gas processing plants and related infrastructure projects completed in 2012 and in 2013. The second quarter throughput was lower than expected due to wet weather conditions that temporarily delayed gathering system construction and producers' well completion efforts. Despite those challenges, we still connected 350 wells in the quarter and 600 year-to-date compared with 450 wells last year through June. While we expect to connect more than 1,000 wells to our Williston Basin and the Mid-Continent gathering systems in 2013, we reduced this segment's 2013 operating income guidance to $209 million, reflecting lower natural gas volumes gathered and processed experienced in the second quarter 2013 compared with previous estimates. Specific volume expectations and updated 2013 price assumptions are provided in our earnings release. The natural gas pipeline segment's second quarter financial results were slightly lower compared with the same period in 2012. Equity earnings from Northern Border Pipeline were lower in the second quarter due to reduced transportation rates on Northern Border Pipeline due to a rate settlement with its shippers that took effect in January, offset partially by higher transportation volumes. Substantially all of Northern Border Pipeline's long-haul transportation capacity…

Thank you, Terry. I appreciate that, that was very good, I appreciate that. Before we take your questions about the quarter, I'd like to take this time to reiterate our expectations regarding the dividend growth ONEOK anticipates, following the completion of its plan to separate the natural gas distribution business. What we stated in last week's news release on Page 3 and in the slides on Page 9 provided in advance of the conference call is correct. In all the to-ing and fro-ing during the Q&A portion of the conference call, there appears to have been some confusion about what our intent is. So let me reiterate our post-separation dividend growth expectations so that there is no additional misunderstanding. When the separation is complete, post spend, ONEOK intends to increase its dividend to a level competitive with its pure play GP peers. That means the dividend growth rate is expected to be greater than the current pre-spend 3-year dividend growth rate ONEOK estimates, which is 55% to 65% during the 2012 to 2015 period. This post spend ONEOK dividend growth expectation excludes the dividend that ONE Gas would pay its shareholders, which we expect would also be competitive with its natural gas utility peers. Said another way, ONEOK dividend growth post spin is exclusive of the ONE Gas dividend. The ONEOK dividend growth expectation post spin comes entirely from ONEOK. As usual, in closing, I'd like to again thank our 5,000 employees whose commitment, dedication, skills and experience allow us to operate our assets safely reliably and environmentally responsibly every day, and to create exceptional value for our investors and our customers. With our announcement last week, all of our employees are facing change. But the approximately 600 employees in our shared service organizations are facing more change and uncertainty because they do not know if they will remain with ONEOK or move to ONE Gas. Removing that uncertainty is a high priority for my leadership team during the next 60 to 90 days. Our entire management team appreciates all of our employees' efforts to make our company successful. Operator, we're now ready for your questions.

[Operator Instructions] Our first question today will come from Christine Cho of Barclays.

If I do some quick back-of-the-envelope math, it seems that your optimization margin was much higher than what some of us were probably expecting and better than what we've seen in the past 2 quarters. Would it be fair to say that year-to-date, you guys are tracking higher than what you would have expected? And how much of your increase in guidance for the NGL segment reflects what you've already generated in the past 2 quarters versus what you hope to see in the next 2 quarters?

Terry?

Well, actually, the impact from ethane from optimization, the revenue impact for 2013 is approximately -- let me see, hang on just a second. Let me make sure I get your question right. Christine, could you say the first part of your question again?

It seems that your optimization margin for this quarter was much higher than what some of us were probably expecting and better than what we've seen in past 2 quarters. So would it be fair to say that year-to-date, you guys are tracking higher than what you would have expected 6 months ago? And how much of your increase in guidance for the NGL segment reflects what you, call it, already have in the bag versus what you hope to generate in the next 2 quarters?

The answer to that is yes. Yes, it is tracking higher.

The thing I'd add though, Christine, you can see that we averaged $0.01 on the spread in the first quarter and $0.06 in the second quarter, averaging $0.04 for the first half. We do expect that to be higher in the second half, as well as the optimization volumes to be higher as well. So it is a little more back-end weighted.

I see. Okay. That answers it. And then, when ethane spreads were very wide last year and the year before, most of your optimization capacity was used for ethane. For talking purposes, let's say it was 100,000 barrels per day for all ethane, if we were to instead use all propane or all butane, am I correct in thinking that this number moves lower because the other liquids are heavier and they take longer to move?

No, I don't think so.

Okay. And then, just my last question, it sounds like with the export projects and propane inventories being relatively low, we might see some strength in propane prices, especially if we get a normal winter. I'm assuming this dynamic is what's driving your lower expectation for ethane rejection if the older plants are losing their propane recoveries with the ethane rejection?

Yes, I think that's right.

Our next question comes from Carl Kirst of BMO Capital.

Maybe just queuing off of Christine's question, and perhaps asked another way on the NGL guidance. If you think about it for a full year, the segment guidance is going up $30 million. Is it possible to segregate, since we have this situation where the fractionation, the transport volumes are being reduced, lower but have some protection by some ship or pay, is it possible to segregate what the negative impact of the reduced volumes, what that impact is versus the uplift from the optimization that gives this net $30 million uplift?

Carl, there is actually kind of more to it than that. You're hitting on the optimization piece. You hit on the minimum volume requirement piece. There's 3 other factors here that really play into this that aren't volume related. The first one is we have more isom volumes available to us to process through our isomerization unit. We have higher spreads on that volume as well. The next piece is we're experiencing more commercial marketing opportunities because of assets that have been freed up by putting the Bakken line into service. For instance, we got more truck and rail activity that used to be dedicated to the Bakken barrels. Now we're getting third-party business through that. And then, we also have the propane export opportunities. And the final thing is there are incremental volumes that are actually coming on, but they're not enough to offset the reduction in some of the gathered volumes. So all that combination of things with the optimization volumes, the minimum volume requirements, is actually making up that $30 million.

Great. I appreciate the extra color. And then, just a second question on the gathering and processing, and it looks like with the guidance reduction that's going on there for that particular segment, we may have, on the gathered, processed perhaps, volumes being down 2% to 4% relative to prior guidance, maybe a little bit more so on the implied equity production. But I guess my question is, it sounds like that's all timing-related with respect to well connects in the second quarter. Is that something where you feel like, by the end of the year, you will have caught up to your prior expectations? Meaning, even though we haven't seen granular guidance for 2014, and sort of what you guys are looking at, you're not really expecting much of a change from the volume or does this push out the entire curve where we might expect lower volumes, both in 2014 and 2015, because we've been backed up by a quarter? If that makes sense?

It makes good sense. It just backs the curve out.

It slides the curve.

Moving on, we'll take a question from Ted Durbin of Goldman Sachs.

I guess, my first question here is on the exchange service activities, and I'm assuming there's sort of a layer of contracts you've signed here on the minimum volume. So I'm wondering what the vintages of those is and how we should think about the change in the contract pricing you signed up for the shippers on there? I'm assuming, as the spreads here between say Conway and Belvieu have come in, you priced on those contracts up at higher rates that maybe [indiscernible] maybe I'm thinking about that wrong?

Well, you're thinking about it right. We're just not going to be willing to tell you the specifics, I mean, for competitive reasons, right? That's exactly right.

Okay. No, that's fair. And then, second one for me was just kind of where you are on hedging for NGLs. It looks like you're still leaving 2014 fairly open. Is that just because you don't like the price -- I mean, you kind of walked through your ethane price and propane price forecasts, is that kind of a big driver there, where the hedges are?

I think, that's right, Ted. When you look at ethane and you look in the out months from a hedging standpoint, you have take a pretty significant haircut in pricing. And what we've tended to do over time, particularly as it related to ethane prices, generally move back towards us. So we're just going to continue to look and look for the opportunities to hedge more ethane as we go in the future.

Got it. That's great. And then, last one for me. Just on the balance sheet at OKS. I think, I heard you say, it was a -- you're 3.6x on debt to EBITDA, I just want to make sure, is that on a trailing 12 months? I'm coming to a higher number, I'm just kind of thinking about where you are in your credit metrics? And then how you're thinking about financing the CapEx for those sort of kind of where you are?

Derek?

Ted, that's correct. Although the 3.6 is adjusted for some EBITDA on our capital projects that we get credit for in our covenants.

Got it. So you're giving some forward EBITDA, I guess, credit for things that are coming on?

That's correct.

We are allowed to do that by our agreement.

Yes. And then, how are you thinking about sort of financing the CapEx for the next 6 to 12 months, let's say?

Well, I think, as we said before, we do have equity and debt needs, and the timing of that, we'll just have to determine based on the market conditions.

Moving on, we'll hear from John Tysseland of Citi.

Can you just describe the mechanics and maybe provide an example of your exchange services operations? And then, kind of what the difference is of that versus maybe optimization? And then also, where geographically you see the most opportunities in today's kind of market, is that -- in that business, is it Bakken or Mid-Con?

Well, first off, when we look at -- we discussed exchange services, that's all fee-based, that's all fee-based business, okay. Transportation, frac services for the gathering fractionation. If we look at optimization, of course, that's spread. It's the Conway to Belvieu pricing differentials for ethane is what we talk about. The other part of your question is, where do we the opportunities? We continue to see a lot of opportunities in the NGL rich plays within our footprint, the Bakken, of course, continues to yield a lot of opportunity. The Niobrara down in the Powder River Basin is yielding some opportunity for us. The Mississippian Lime down in Oklahoma and the Cana-Woodford play are all plays that are still very active and new processing plants are being developed.

That's helpful. Can you just describe the mechanics of how that contract with customers actually works from -- in the exchange services business? I mean, what is the producer -- what is the contract with the producer or the downstream customer really have with ONEOK?

Well, basically, how the exchange services contract works, as I said before, it's a fee-based contract. We take the barrels and exchange raw feed for purity products at the market hub. And we do that for a fee. Okay. So that's, in a nutshell, how it works. The transportation or gathering and fractionation fees are typically bundled, okay. So you get all of that for a flat fee. So that's basically how it works.

So, to put numbers to it, just using 1 commodity, we take the y grade -- we take the raw, right, we take title to it, say, in the Bakken. And then, we provide a price, and then we charge the producer based on the finished product at the market hub, like Terry said, less a fee. So in other words, we do the gathering, we do all the transportation, we do the fractionation, we distribute the products to the market hubs, we -- and then give them, in essence, market price, less a fee, for doing all that, all those fees, we don't -- we've got 1 fee. So that's why we call it a fee-based business.

That makes a lot of sense. I appreciate the clarification. And then, final follow-up on that same kind of line of questioning. When the Bakken pipeline actually came online, is it fair to say that your exchange services business and opportunities there expanded pretty greatly during the quarter?

That would be right.

Our next question comes from Craig Shere of Tuohy Brothers.

Look forward to seeing you guys next week. Just a couple of follow-up questions on the optimization. Seems like the ability to have some optimization fill in for lost capacity utilization amidst ethane rejection is suggesting that there's a lot more opportunity other than the historical Conway to Belvieu ethane spreads. And I think, on Christine's questioning, the answer was that this year's optimization, even though there were some pretty healthy results this quarter, were more back-end loaded towards the second half. Could you comment about the various opportunities for optimization and the sustainability of that into 2014 and beyond, as ethane rejection continues, albeit at a lower rate?

Well, Craig, we don't provide the specifics as far as volume for our optimization activities. But the opportunities for optimization span across a number of products, not just ethane, propane, normal butane, et cetera. So we optimize more than just ethane. We're going to continue to see a need for products in Mont Belvieu, we're going to continue to see growth in supply in the Mid-Continent, based upon what we're seeing in terms of development in NGL rich plays. So I think the opportunities for optimization as a viable portion of our business are going to continue, although they're going to be a much smaller percentage of our margin versus what we had seen over the last couple years.

[indiscernible] You were knocking the lights out a year ago, but I think what we're seeing now is definitely better than the market was expecting, at least the financial markets.

Right.

If I could jump over to...

One of the things I would add to that, knocking it out of the ballpark is remember that the -- it's the optionality to the asset that allows us to capture the optimization margin, regardless of what the margin is. We don't create the margin. It's the ability to capture it, whether it's $0.01 or whether it's $0.40. It's the optionality embedded in those assets and our ability to bundle all those services together, relative to our competitors, that gives us an advantage to participate in optimization. That's an important piece.

Understood. And then, that's a nice segue for my follow-up. Obviously, over time, the desire is to convert more and more towards fee-based exchange service activities, and I think you all had alluded in prior commentary to expectations that you could contract out multi-year agreements on Conway to Belvieu spreads capacity at close to $0.10. In light of some of your more recent commentary around the ethane markets, could you give any colors, do [ph] the market's appetite for multi-year agreements fee-based that you would be comfortable with or is that not available in the short run?

Well, our strategy has been to long-term contract that optimization capacity under fee-based exchange services. As we look at the marketplace -- and our customers, they want that long-term capacity. And the fact that optimization spreads or the Conway to Belvieu spreads have been very volatile and basically nonexistent in the first quarter, has not fazed our customers one bit, they still want that capacity, they want to contract for it long-term, and they're willing to pay an attractive rate or a fair rate to do that.

Well, I guess, let me rephrase, instead of specific pricing like $0.10. We saw a nice uptick, even towards $0.07 plus until that unfortunate olefins plant accident. And then we gave back $0.01 or $0.02. Do you see your customers looking past the near-term issues towards multi-year -- the desire, as you to say, to lock in multi-year spreads at sustainable levels comfortably above what we're seeing now, which is more like $0.05?

I don't -- and here's just my opinion. These customers are not focused on the spread. What they're focused on is getting their barrels to where they think the best markets are. So they haven't really been fixated on the spreads and they don't come back to us and say, well, we think the spreads going to be x, so as a result, we think your rate should be y. They recognize the cost it takes to build infrastructure down to the Gulf Coast, they want to be in that liquid market, because they believe that's the place to be. So that's where we're taking them. So we don't really get into that issue with the producers arguing over what the view of the spread is.

Is a part of it that you simply don't get in this argument because you provide so much bundled services that it's simply one part of the total?

I think that's definitely key, that we're able to provide that full menu of services relative to others in the industry. I think that's very important.

I think it probably is the key, that's what differentiates us from others.

Up next, we have Jason Agnu [ph] of Robert W. Baird.

With plenty available revolver capacity and potential to increase that base, are you guys solely focused on directing that capital towards your backlog of organic projects or any M&A opportunities, I guess, outside of your current fairway that you're evaluating?

Terry, do you want to take that?

Well, I'll take it. The answer is yes, we will continue to look at M&A activities and then we'll look for an opportunity to finance those M&A activities. As I've said many times before, we have and will continue to look at opportunities to grow through acquisition. But as I've also said many times before, it's a lot more economical use of our capital to build, at this point in time, than to buy, when we measure these opportunities against our next alternative.

Got it. And then more of a housekeeping item, but maintenance CapEx seems to be weighted a little more towards the back half of the year. Is that something we should expect kind of going forward into 2014 and 2015?

Yes. This is -- as you would imagine, with more assets coming on stream, you're going to have smaller incremental amounts of maintenance capital.

Moving on, we'll hear from Matt Niblack with HITE.

Not to add confusion to the distribution growth question, but I just wanted to clarify, when you talk about that growth rate being higher than it was before exclusively for the remaining ONEOK, is that based on a baseline of the current combined distribution, or is that based on a new baseline of the distribution as it will be after the separation?

Okay. Let me come back and repeat what I talked about earlier. When we talk about -- when the separation is complete, after the spin of the gas utility, ONEOK intends to increase its dividend to a level competitive with its pure play GP peers. If you look at the pure play GP peers and measure that against where ONEOK is today, the pure play GPs are -- show a greater increase, okay? Because that's what we're saying. Did that help you?

Right. A greater growth rate, right?

Yes.

Okay. So when we think of that greater growth rate...

Whoa, I'm sorry to interrupt you. But also, what I said was ONEOK intends to increase its dividend to a level competitive with its pure play GP peers. Now that's what I said, and those words, to me, mean that ONEOK intends to increase its dividend.

Right. The remaining ONEOK will increase its dividend versus current levels in absolute terms?

That's correct.

Our next question will come from Chris Sighinolfi of Jefferies.

Terry, I just have 1 question remaining, and that's pertaining to the NGL sales for the quarter. Obviously, they jumped quite significantly, you mentioned that in your prepared commentary. I just wanted to better understand the drivers behind that, obviously, you had the ethane header at Belvieu and some other things. But I'm just, I guess, curious, was inventory sales a contributor to that jump at all? And if so, roughly how much?

Yes, it definitely related to inventory. The reason why that sales number has increased is we're loading our storage down there at Belvieu with raw feed to get positioned and ready to start up the MB-2 fractionator. So in all of the accounting associated with that, in protecting the price risk associated with that storage, it inflates our sales number. So that's what you're seeing. It's not related to any sort of contractual change or anything like that.

Okay. So with that facility slated to come on here in 3Q, can you quantify at all, like how much that impact might have been in the quarter, in 2Q?

Related to that -- the inventory?

Yes.

Probably, that delta, the delta Q-over-Q is probably all attributable to that raw feed build.

We'll take our final question from John Edwards of Crédit Suisse. John Edwards - Crédit Suisse AG, Research Division: Just if we could come back to the ethane rejection versus optimization and the exchange services issues. Just first on exchange services, kind of a follow-up to John's question. So the margin, does that vary with respect to market prices and volumes involved in the exchange? And if so, can you give us any kind of idea how that would work?

Think of it this way: Optimization is going to be the relative value between 2 market points, Conway and Belvieu. Less whatever cost might be incurred, you move the product between the 2 points. Now to avoid the cost, you could buy a product at 1 point and sell it at another, et cetera. But optimization is all about just taking advantage of whatever, if you will, spread, is I think the word Terry uses, that exists between a relative A component of the NGL barrel between 2 locations at a particular point in time. And that is just -- and so what -- and you compare that to an exchange service fee, and as Terry mentioned earlier, when we work with our producers, the optimization piece is not part of the puzzle. The fee represents a fair exchange of cost, or revenue in our case, cost in the case of a producer, for us to gather the NGL barrel, bring it to our fractionation facility, fractionate it, distribute the finished products to a market center, and then monetize that product for the producer in the form of a price, less a fee. And so there's no spread component in the exchange services agreement. John Edwards - Crédit Suisse AG, Research Division: Okay. That's helpful. Just -- and then adding to that, is that fee fixed that you're talking about to provide all those services, or does that fee itself also vary depending upon the volume that you're doing it for and the market price or how does that vary?

It varies by the quantity, the relative terms that the producer wants. Obviously, a producer that has more barrels is probably in a better position of negotiating with us than one that has fewer. But that's an example. But each of those contracts, although there's a fee, also has, sometimes, the ability for that fee to change based on increased quantities or actually reductions in quantity. And then, it also will include things like consumer price index, power price indexes, things of that nature, that the 2 parties agree to, to adjust the fees, so that neither party suffers relative to increased or decreased cost. And all of those things are negotiated in these exchange service agreements. And that's how we're able to get the term, is by reducing the amount of exposure to both parties, finding that good middle ground. John Edwards - Crédit Suisse AG, Research Division: Okay. And that fee, is it basically on a -- I mean, I realize there's a lot of services here, is it the form, is it on a per barrel basis?

I think it's on per gallon but... John Edwards - Crédit Suisse AG, Research Division: Per gallon.

It can be on a per barrel or gallon basis, whatever you choose to do. John Edwards - Crédit Suisse AG, Research Division: Okay. And then, just to help us for modeling purposes. Is there kind of a range that we should be thinking about? Or is that something you can't talk about?

We really can't talk about that.

I would say this though. We can't -- the piece in the marketplace that's transparent is the optimization spread. That's out there every day. The exchange with the producer is providing all those services and getting them, in essence, finished barrel, finished product pricing less the fee agreed to, to make all that happen for them. So if you pick Belvieu as the point in which the pricing under the contract is agreed to, it's going to be Belvieu pricing less the fee on the exchange service agreement, for simplicity's sake. And that is the producer's net back will be reduced by the fee that they pay ONEOK. And that fee can be adjusted, and we talked about some of those adjustments. Now what's happened is, as ethane has come off the system because of ethane rejection, it has created additional space in all of those pieces of pipe, whether it's gathering, whether it's fractionation, whether it's distributed pipelines, right? What ONEOK has been able to do for its own account is to either buy barrels at -- or sell barrel at Conway or buy barrels at Belvieu or sell barrels at Belvieu, buy barrels or move them back and forth, paying -- covering those costs and making money with that space that's been created by the fact that ethane is no longer being produced by that gas processor. So the ethane play on spread play is for our book. I mean, for the most part, that's the way to think about it. And then the other part is just an exchange of service for the -- with the producer for the services that we provide them and it removes, to a very large degree, removes any price risk. John Edwards - Crédit Suisse AG, Research Division: Okay. That's really helpful. And that actually is a good segue then to the question. I was trying to understand a little better that interaction between the ethane rejection and uplift from optimization. I mean, in the guidance that you gave, you went from $5.45 to $5.75 for the NGL segment. You were citing the optimization versus ethane rejection. Is there any way to sort of break that out? How much was the downshift from ethane rejection? How much was the uplift from optimization?

Well, what I... John Edwards - Crédit Suisse AG, Research Division: The net $30 million, how much was uplift from optimization, how much was lower from ethane rejection, I guess?

Let me try this. Without -- we don't -- as I've indicated to you before, we don't provide optimization volumes. But if you look at our guidance for 2013, and we're -- we've got ethane rejection in that guidance, okay? We're offsetting a considerable portion of that ethane rejection with the optimization activity. But if you look at optimization activity, to provide some relativity for you, if you look at the optimization margin that we expect to generate in our 2013 guidance, it's going to be in the range of what we have indicated we're going to be running, 15% to 20% of our margin, okay, will be from rejection. The rest of these things, minimum volume obligations, additional Bakken volume, significantly higher isom margins, various term transactions and commercial transactions in our marketing, those are all things that are going to contribute to that guidance. And optimization is that -- is in that -- is not an unusually large percentage.

So when confronted with the fact that ethane comes off the pipe, if you will, or comes out of the fractionator, so you've got this empty space, what are you going to do with it? Well, as I alluded to earlier, the optionality that the assets give you is to go out, because you have all these interconnected assets, the gathering, the fractionation, the distribution, the storage, you go out and you make the most out of all those assets and take place of what the -- take advantage or capture what the market gives you, and that's the optimization. The alternative is to let all these other things happen that Terry's talking about, the isomerization activities, the increased volumes that we've gotten from the new supply connections, et cetera, new pipeline out of the Bakken, you take all those things and those are all great things and they all added to and have added to and will add to our bottom line. But in addition to that, instead of just letting the space go empty, we've taken advantage of the optionality that the assets provide and capture some value for that. And as Terry's alluding to, it's not -- I mean -- we throw these millions of dollars around like they're nothing, but it's a decent amount of money, but it didn't turn the whole tide. It's not what puts us where we are today, the other stuff. Did that help you? John Edwards - Crédit Suisse AG, Research Division: That's helpful. Maybe I'll follow-up a little bit further off-line.

That does conclude our question-and-answer session. I'll turn the conference back over to our speakers for any additional or closing remarks.

Well, thank you for joining us, everybody. Our quiet period for the third quarter starts when we close our books in early October, and extends until earnings are released after the market closes on November 5, followed by our conference call at 11 a.m. Eastern time, 10 a.m. Central on November 6. We'll provide details on the conference call at a later date. T.D. Eureste and I will be available throughout the day to answer your follow-up questions. Thank you for joining us and have a good day.

Again, ladies and gentlemen, that does conclude today's conference. We do thank you all for joining.