CF工業控股 (CF) 2017 Q2 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Second Quarter 2017 CF Industries Holdings Earnings Conference Call. My name is Cynthia, and I will be your coordinator for today. (Operator Instructions) .

I would now like to turn the presentation over to the host for today, Mr. Martin Jarosick, with CF Investor Relations. Sir, please proceed.

Good morning, and thanks for joining us on this conference call for CF Industries Holdings Inc. I'm Martin Jarosick, Vice President of Investor Relations for CF. With me today are Tony Will, our President and CEO; Dennis Kelleher, our Senior Vice President and CFO; Bert Frost, our Senior Vice President of Sales, Market Development and Supply Chain; and Chris Bohn, our Senior Vice President of Manufacturing and Distribution.

CF Industries Holdings Inc. reported its second quarter 2017 results yesterday afternoon, as did Terra Nitrogen Company, L.P. On this call, we'll review the CF Industries' results in detail and discuss our outlook, referring to several of the slides that are posted on our website. At the end of the call, we'll host a question-and-answer session related to the company's financial results for the quarter. As you review the news releases posted on the Investor Relations section of our website at cfindustries.com and as you listen to the conference call, please recognize that they contain forward-looking statements as defined by federal securities laws. Actual results may differ materially from those projected as a result of certain risks and uncertainties, including those detailed on Slide 2 of the accompanying presentation and, from time to time, in the company's Securities and Exchange Commission filings. These forward-looking statements are made as of today and the company assumes no obligation to update any forward-looking statements.

This conference call will include the discussion of certain non-GAAP financial measures. In each case, the presentation of the most directly comparable financial measure calculated and presented in accordance with GAAP and coinciding reconciliation of the non-GAAP financial measure to the most directly comparable financial measure calculated and presented in accordance with GAAP, is provided in the earnings release and the slides for this webcast presentation on the company's website at cfindustries.com.

Now let me introduce Tony Will, our President and CEO.

Thanks, Martin, and good morning, everyone. Last night, we posted our financial results for the second quarter of 2017, in which we generated adjusted EBITDA of $303 million after taking into account the items detailed in our earnings release.

The story of the quarter was exceptional execution by the CF team. Record production in sales volumes, outstanding safety, good price realizations and cost reductions against the backdrop of difficult market conditions. The global market was impacted due to an excess amount of traded material. Low-cost, export-oriented production has increased while at the same time, other regions have also brought on new production, decreasing their import requirements like in North America. This has led to an excess of available tons and resulted in sloppiness in the market. Although we fully expect that some higher-cost producers will be forced to curtail or permanently shut capacity and for product flows to adjust, it will take time for the global trade flows to realign.

We do see some encouraging signs that this is beginning to happen, as we have summarized on Slide 13. However, there is still a ways to go. China is supplying less nitrogen to the international market because global prices are below the cash cost of many Chinese producers. Urea exports from China this year through June totaled 2.8 million metric tons, a decline of 46% from the same period last year. In February, we projected that China would export between 5 million to 6 million metric tons of urea in 2017, with operating rates at approximately 60% of capacity, actual exports may, in fact, be below our estimate.

Exports are also down from other marginal producers in Eastern Europe, South Asia and Ukraine. Based on public announcements, we estimate that more than 10 million metric tons of ammonia capacity and 8 million metric tons of urea capacity in these areas have been curtailed for extended periods over the last year. Although we are seeing improvements like those, more is needed.

In the quarter, certain producers and traders with excess tons needed to unload positions and find liquidity and unfortunately, North America was the dumping ground. North America urea and UAN imports through May were down 22% and 10% year-over-year, respectively. However, these levels were still substantially higher than what is required, given new North American capacity. Consequently, urea barge prices at New Orleans averaged $30 per ton below international parity during the quarter and broke through 2016 lows in June.

Even with NOLA barge prices so low, shipments continued to arrive in the second quarter. It is clearly uneconomic to force more tons in the North America than what is consumed here, driving prices below international parity, only to have those same excess tons reexported again later. Product movements like those are part of the trade flows that need to rebalance in order to eliminate obvious inefficiencies in the global system.

Looking forward, we project that nitrogen prices will be challenged through the remainder of 2017 and into 2018. Global energy prices remain low and still more production capacity is yet to come online. Despite modest urea price increases we have seen since the end of June, the third quarter typically has the lowest demand and prices of the year for all products, as you can see on Slide 16. If substantial imports arrive during the quarter, we could see pricing test to new lows. However, our longer-term outlook remains very positive and is well-grounded in the industry fundamentals. Post-2018, we project the rate of net new capacity additions will be well below expected global demand growth, which should help pull the industry out of the cyclical lows we are experiencing today and lead to sector recovery over the next few years.

Now I want to highlight some of our outstanding execution in the quarter. We ran our plants really well. We produced an all-time record 2.7 million tons of ammonia while doing it safely, achieving the 12-month recordable incident rate of 0.91 incidents per 200,000 hours worked. This is especially remarkable considering the number of new operating units and the number of new people we have added in the last 2 years. The new Port Neal ammonia plant ran at more than 115% of nameplate capacity during the quarter and we have 10 other ammonia units that ran above 99% of their stated capacity. This was truly a systemwide accomplishment.

Our sales and logistics team took that record production and turned it into record sales volumes. They did a great job of navigating challenging conditions and moving tons to the right place at the right time. In some cases, the right place was offshore. During the quarter, we used the destination optionality, we created to export nearly 100,000 tons of urea from Donaldsonville. These exports averaged more than $10 per ton, higher than prevailing NOLA urea barge prices at the time. We also took advantage of export opportunities for more than 200,000 tons of ammonia and 300,000 tons of UAN. As a result, we proactively managed our system inventory, kept our plants running full and realized good overall cash margins for the business.

With that, let me turn the call over to Bert, who will talk more about the commercial environment during the quarter and looking forward. Then Dennis will discuss our financial position. Additionally, we're going to try something a bit new for us here at CF. Dennis is going to highlight some key market data points and provide some handrails for people trying to model our third quarter performance in an attempt to help them calculate a more accurate result, so I encourage you to listen to what Dennis has to say.

Okay, with that, I will now turn the call over to Bert.

Thanks, Tony. CF's performance in the second quarter demonstrated the strength and agility of our team and system. We employed our storage assets, shifted our product mix as needed and sent product to twice as many countries as we did in the second quarter of 2016. As a result, we set company production and sales volume records and maximized the overall margin available to us despite the challenging business conditions.

New capacity and low energy cost worldwide continued to pressure nitrogen prices. As Tony said, the market is in transition and global trade flows are adjusting. We expect that this process and its effect on pricing will continue through the remainder of 2017 and into 2018. Low nitrogen prices are affecting most of our segments, but we saw localized factors play a role in our performance as well during the second quarter. Wet and cold weather in North America that lasted from mid-March through April disrupted the spring application season for ammonia, urea and UAN.

For ammonia, this resulted in a lower application totals as some demand moved to upgraded products. We moved the balance of ammonia, we had expected to sell, to agricultural customers as well as the incremental tons produced from our expanded capacity into the export channel. Export sales were higher than expected as a result and lowered our average sales price for ammonia during the quarter. We expect our ammonia exports to decline significantly by the end of the year when Mosaic's vessel is ready and they begin taking ammonia delivery from our Donaldsonville facility according to our supply agreement.

Turning to urea. High imports in the first quarter led to a price reset, and although it was more pronounced and earlier than expected. The urea barge prices at NOLA fell 38% from the Q1 high to a low of Q2 as weather slowed demand and distribution channels in North America already had sufficient inventory. With the NOLA barge price, $30 below international parity, we are able to take advantage of export opportunities at a premium to NOLA barge prices, as Tony mentioned earlier. We believe the price disparity between NOLA and the rest of the world is not sustainable and that the urea pricing will converge as trade flows adjust. While we produced record levels of ammonia and urea this quarter, we made less UAN than the same period the year before, as we adjusted our project mix to reflect better opportunities with other products. The wet and cold weather led to late planting and slower growing conditions, delaying UAN purchases and applications in North America. We were able to use our storage assets to capture the resulting late-season demand that stretched into July. We've seen continued volume momentum with UAN, as we launched our fill program last week at the equivalent of $125 per short ton at NOLA or about $15 to $20 less per ton than last year, depending on location.

Ammonium nitrate continue to be a bright spot for volumes and pricing due to a long-term supply agreement that commenced in 2017 and additional AN volume produced and sold in the U.K. We continue to grow our sales of Diesel Exhaust Fluid. We expect DEF sales in North America to exceed those in Europe for the first time ever this year. Our new DEF unit in Donaldsonville gives us additional capacity to meet this growing demand and additional product mix flexibility to adjust quickly to changing nitrogen market conditions. We expect the global nitrogen market environment to remain challenging. We're in the midst of the quarter where we typically see our lowest prices and sell the least volume. There's additional capacity in North America coming online over the next few quarters, to which trade flows need to adjust and that will take time.

Buyers in the region remain risk-averse, but more purchasing in a just-in-time manner or selling and purchasing back-to-back. They also appear to be keeping lower inventories through the year. All of these factors on top of the supply-driven global market will restrain prices through the remainder of 2017 and into 2018. We are well prepared for this challenge. We will continue to work with our customers and seek to deliver good overall margins for our business by moving tons to the right areas at the right time, utilizing storage, adjusting product mix and developing customers and new destinations. We have demonstrated our ability to do just that and are well positioned as the global nitrogen market continues to transition.

And with that, I'll turn the call over to Dennis.

Thanks, Bert. In the second quarter of 2017, the company reported fully diluted earnings per share of $0.01 and EBITDA of $275 million. After taking into account the items detailed in our press release, our adjusted earnings per share for the second quarter was $0.10 and adjusted EBITDA was $303 million. Our adjustments include an $18 million unrealized mark-to-market loss on natural gas derivatives.

As you've seen in recent quarters, the number and size of adjustments have decreased significantly compared to last year, as we have reduced activity and completed our capacity expansion projects. We also restructured intercompany loans to reduce the impact of foreign exchange rate changes on these loans. These noncash effects were significant in the first half of 2016, but not in 2017.

Our balance sheet remains strong. Our cash and cash equivalents were $2 billion as of June 30. At the end of the quarter, our $750 million revolving credit facility was undrawn. As we disclosed in June, we received $815 million in federal tax refunds due to the carryback of federal tax losses in 2016 to prior periods. These tax losses were driven primarily by federal tax bonus depreciation on our capacity expansion projects. We intend to use the proceeds from these refunds to retire $800 million of debt coming due in May of 2018. Once this occurs, our quarterly interest expense will decline from $80 million to approximately $66 million.

Depreciation and amortization expense was $217 million for the second quarter of 2017, up from $181 million in the second quarter of 2016. We expect full year 2017 D&A to be approximately $875 million. As we've discussed before, higher D&A this year is affecting gross margin per ton for all the applicable segments. We have again included segment tables in the press release and a chart on Slide 9 to help you compare the gross margins in the second quarter to prior periods. In these tables and charts, we adjust depreciation and amortization and unrealized mark-to-market gains or losses on natural gas derivatives from gross margin as reported to show an adjusted gross margin. We believe this will give you a better sense of the underlying cash generation capability of each segment.

Capital expenditures for the second quarter of 2017 were $91 million. For the full year, we expect to spend approximately $400 million for new activities. Additionally, as of June 30, 2017, approximately $175 million remains accrued, but unpaid related to capacity expansion activities in 2016. Most of this unpaid amount is a subject of dispute to the certain contractors and vendors and it's too soon to estimate the timing or final resolution of these matters. I want to highlight the cost reductions we've achieved across our system. This is the result of our targeted efforts to reduce cost in our head office and at our plants.

Our SG&A is down slightly compared to the same period last year, which is a meaningful achievement considering our increased production volumes and ever-increasing regulatory and reporting requirements. We have the lowest SG&A as a percentage of sales in the fertilizer space and among the lowest in commodity chemicals generally. At our manufacturing sites, reduced plant level activity and better procurement practices have decreased professional services, payroll and maintenance costs, resulting in a lowerable controllable cost of sales. As you can see on Slide 10, we have reduced controllable cost of sales by $28 million compared to the second quarter of 2016 or about 15% per product ton. Year-to-date, these efforts have lowered controllable cost of sales by $45 million.

Now I would like to provide some specific thoughts regarding our expectations for the third quarter of 2017. We expect to see higher volumes and lower controllable costs compared to the third quarter of 2016, due to our expansions and the cost reduction activities, I just mentioned. But we believe these positive impacts will be more than offset by lower product pricing in the market today as well as higher gas costs compared to the third quarter of 2016. As Bert mentioned, we launched our UAN fill program at $125 per short ton NOLA, which is about $15 to $20 lower than the last year. Tampa ammonia prices are $190 a metric ton today compared to about $280 a year ago. NOLA urea barge prices have recently moved up to around $185 per short ton, similar to last year. If you compare natural gas prices for the third quarter of 2016 to the third quarter of 2017, Henry Hub prices are up about $0.25 per MMBtu and NBP prices are up about $0.80. Based on our typical consumption and factoring on our hedge position, this is about $30 million natural gas headwind on its own. Taking all of these factors into account, we believe it is unlikely that our financial results for the third quarter of 2017 will exceed the $83 million of adjusted EBITDA reported in the third quarter of 2016.

With that, Tony will provide some closing remarks before we open the call to Q&A.

Thanks, Dennis. Before we move on to your questions, I want to recognize the winner of our 2017 Stephen R. Wilson Excellence in Safety Award, which was created to help us identify, honor and share the most impactful safety innovations and improvements made across our company every year.

This year, our winning facility is the Courtright Nitrogen Complex for their innovative implementation of technology that eliminates the risk of high-voltage arc flash incidents during infrared scans of our electrical equipment. They've also been nominated for an Ontario Electric Safety Award for the idea. We are very proud of them and how this will improve safety across our system.

I want to thank all of our employees for their work during the quarter. They operated safely, they operated well and they operated as a team. Their execution is even more critical during the current challenging market environment and has positioned CF well for a long-term success.

With that, operator, we will now open the call for questions.

(Operator Instructions) And our first question comes from the line of Chris Parkinson from Crédit Suisse.

So clearly, the market is under pressure, which is dominating the headlines. But I think most will be happy to see you're not sitting idly at all. Just putting the market aside for a second and your views, can you just walk us through a little more on the controllable per ton costs? The end-unit versatility of your assets, especially given your new DEF unit as well as any transportation, logistics benefits inherent in your strategy and how you think these should help you further weather the storms, so to speak, for the back half and into 2018?

I'm going to ask a combination of folks to help answer that question starting off with Chris Bohn on the controllable -- on the manufacturing side.

Yes Chris, on the controllable cost decrease that we saw, it's a makeup of a combination of factors. As you mentioned, the increased volume counted for about 60% of that decrease of the 15% or roughly, as Dennis have mentioned, $15 per ton. But in addition to that, we had lower maintenance and also lower labor costs, as we put focus back on safety and reliability, where in the past, we had some of our focus by our engineering team largely put towards improvement projects. And as we've seen, those projects sort of tail off here. We've been able to reduce cost considerably by that. But additionally, the plants operating as well as they have with the utilization over 100% in the quarter. That also really results in lower maintenance. And it allows us during that time frame to have more planned scheduled maintenance where there's lower cost related to that as well. So this is something that we've increased the focus with all the general managers at the sites. We expect to continue the focus on that and probably lower activity by our engineering group on improvement projects.

Bert, you want to talk about logistics?

So there are a lot of things we're doing just to, as you say, weather the storm. We just view it as optimizing our system and allowing ourselves to have ultimate optionality regarding markets we participate in, geographic areas, exports, storage and so we're maintaining our inventories low. And we entered the fill season with an adequate inventory and we're exiting that kind of sales campaign in a good position. We've got a good order book on with lower inventory in the system, which we can utilize and, as we mentioned in the prepared remarks, the options for exports on diesel is how we balance logistically. What has become difficult, I think, for our industry is the rail rates. As rail rates have maintained high, specifically the UP and the BNSF into some of the Midwest, that has limited some options. So we've been -- become better and we will continue to improve the utilization of trucks throughout our system and serving our customers in that way. So we're always trying to look at our system, how we can improve it and how we can maximize. And I think a great example is the mega tow that we used to haul the UAN up the rivers and lower that net cost to the company.

And just a quick follow-up. Can you just comment...

And our next question comes from the line of Andrew Wong of RBC Capital.

I'd like to ask about the sensitivity table of EBITDA sensitivity on the presentation, which is really helpful. The table I think is based on 2016 pricing relationships between urea and other products like UAN and ammonia. Have any of these relationships changed over the past year or are expected to change? Just looking at prices of urea, for example, we're at $180 to $190 and UAN is at $120. It's pretty big gap versus historical levels. So I mean, commentary around that will be really helpful.

This is Bert. On some of those -- they're changing. We update this chart, but those products are changing, I wouldn't say daily, but they're changing with regularity, as demand periods are being utilized, for example, ammonia is a fall application and as well as spring, but UAN is principally a spring application and urea is more throughout the year. And so as trade flows and as products come in, some products are weaker sometimes than the others, but generally, the historical spreads do hold and UAN has traded at a premium to urea and ammonia and urea at a premium to ammonia. So it's kind of stairsteps and we use that in a broad base.

The other thing that I'd add to that is that what we will do typically when we will provide projections to the management team here and also to the Board of Directors, we will sensitize -- we will sensitize our projection, not what's shown here on this table, but our projections that we do internally to whatever the latest view is on product premiums and what have you in different locations. And as we've done that and as those things change through time, what we have found is that this table remains relatively robust. As we think about sort of where our projection is and we look at the average gas pricing and the average realized urea price, it kind of -- it remains relatively robust to the changes that we ourselves made to sort of the differences between the products.

Okay. That's very helpful. And then maybe just a follow-up on the longer-term S&D. Looking at your capacity additions, there are about 30 million tons of new capacity being added through the 2020, market growth is sort of about 3 million to 4 million tons. I read your commentary that capacity additions will slow down in 2019. Is that maybe when prices start getting a little bit better? And we start seeing more meaningful increase? Or does it take, maybe, a little bit longer to absorb that new capacity?

Well, Andrew, I think, one of the things that we would expect, there's a difference also between the kind of the gross amount of new capacity coming online and what's not. So the one thing that I want to highlight is there's an awful lot of capacity that's also shutting down, either because it's uneconomic to run at current pricing or because feedstock availability is challenging or a combination of factors. And so what's really important to look at is kind of what the net add is in each period versus the growth rate. But yes, we would expect things to kind of get back to about level in '18 and then start improving in '19 and then accelerate the improvement as you get into '20. Part of kind of that is predicated on our belief that you can't have sort of an infinite amount of capital that's lost and -- across the global system and then a lot of these regions, there isn't enough cash being generated to go through turnarounds on the operating units. And so that's basically when they start shutting down. So it certainly could happen sort of earlier than that, but at least the way that the data is playing out, we would expect kind of recovery to happen in sort of '19 and then accelerate as we go forward. That's a little bit later than what we had said a call or 2 ago and the principal reason for that is a lot of these projects are just delayed in terms of their start-up. And so they cascade kind of out forward instead of starting up when they were originally scheduled.

And our next question comes from the line of Joel Jackson from BMO Capital.

We've seen at NOLA, ammonia prices are now trading pretty much in line with urea, maybe, at a small premium. Historically, when that premium has been near 0 or just above 0 for ammonia at NOLA, you've seen a sharp rebound in ammonia prices. And pretty quickly. Now is that possible this year, considering all the new capacity coming on?

I believe it is. I think, when you look at ammonia and the discount that it's trading at is not sustainable. I think that's a reflection of 1 or 2 vessels that were traded. And you look at the production costs for most people around the world into supply, ammonia delivered at 190 metric ton means and that will be a number that's probably close to production cost, again, depending on your gas cost. And so -- but how we're looking at ammonia, because it's -- we're principally focused on the add market and the demand, the ability to get that product up and through, the ability to supply on a 2 to 4 a week window, the separation between Tampa and the Midwest, I think, will become more pronounced if Tampa stays at this level because that does not reflect an ag value for ammonia applied to the ground.

Okay. That's helpful. So my other question in talking about some of the commentary around Q3. Of course, NOLA is up to $185 or $190 supposedly last night. It averaged -- for urea, excuse me -- it averaged $180, the benchmark did Q3 of last year, your realized premium was only about $22, which isn't what you had in the past. So when you talk about your commentary that you might see lower prices in testing lows -- like testing lows, can you elaborate on that a little bit more? Because again, we're seeing a rebound and to guess it has flows, you have to see a very sharp decline and quickly.

So Andrew, one of the things about -- Joel, I'm sorry. One of the things about the premium last year in market is that we didn't have the Port Neal urea plant up and operating. So with more urea capacity sort of in market as opposed to it being more NOLA-centric, that definitely helps from that perspective. But the challenge, as you know, there's very little product that is going to ground from an application standpoint from this point on until you get into the next application window in the spring, at least, in the northern part of the Cornbelt system. And so urea that is being purchased is basically going into shed for the next 9 months. And so if you end up with an extraneous vessel or 2 that shows up over here when it's not needed, particularly with some of the other capacity coming online, that can have a pretty outsized impact in terms of what local pricing is. And particularly, with a bunch of traders and other folks having a short leash in terms of their open book in the risk positions, then that stuff just starts trading and it creates a cascade effect in terms of price. So that's the cautionary word out there. It's not where price is today. It's the fact that all it takes is 1 extra vessel to kind of rain on the parade.

So really, the message to importers is stay away. Is that right?

Well, I mean, I guess, the message to importers are, there's probably other better value places in the world from a destination market perspective. Because if you come here and you crater the market by $30, you're not doing yourself any favors.

And our next question comes from the line of P.J. Juvekar from Citibank.

It's Dan Jester on for P.J. So, looking -- I want to go back to one of your other questions about the supply-demand balance. In 2015, 2016, you got about 12 million or 13 million tons of urea capacity getting shut down. But this year and next year, you only have 2 million or 3 million tons. So can you just walk us through the factors why, even though market conditions, maybe gotten worse, maybe, not seeing as many permanent shutdowns as you saw in the past?

I think, a lot of it just has to do with in terms of where the hydrocarbon spread is. You don't see as many people kind of -- it's getting to a fairly flat plateau at the top-end and you don't see as many people sitting well off the curve, they're clearly upside down. So when you get to a plateau period like that, then ultimately, what's happening is folks have to sort of -- they go through a period of time where they're not making cash, not hemorrhaging it, but not making it. And then when the next big turnaround cycle comes along, it's typically when they're in the decision around shutting down. And so that's the expectation.

I think, it's also reflected in operating rates. And as the industry's operating around 80-plus percent, you don't have 100% of the static capacity available. And as -- Tony is right and I agree with him that as prices remain low and cash is not available, that operating rate will continue to decline unless overall product will be available, irrespective of the new capacities coming on.

Okay. That's great. And then Bert, I think you touched on this briefly in your prepared remarks, but can you just comment on what you're seeing in terms of the U.S. dealer inventory level exiting the spring and going into the fill season?

We think it's very low. We think it was low as reflected in what we were able to accomplish in June and July, maintaining UAN prices at market levels and having consistent demand every day. Because if the inventory -- the first desire for that retail chain is to be out of inventory at the end of the year. And so that generally happens earlier than they anticipate just because they don't want to carry it. And then we maintain our inventory position in the interior and continue to supply spot sales or new orders up to the day we launched our UAN fill program last week. And so that was for UAN, urea, and then earlier in the quarter, we launched ammonia fill program where these smaller bullets and smaller tanks, it's not a big program, but the industry participates in that where you fill that up at one time and that was completely empty. So we feel very, very positive going into Q3 and Q4 that we'll continue to grow and as demand grows, and we can supply that product and fill those sheds for spring.

And our next question comes from the line of Don Carson from Susquehanna.

This is Ben Richardson sitting in for Don this morning. So I wondered if you could talk a little about your export strategy going into the back half? Do you plan to export more than 2Q?

Just, sort of, that is very high level as Bert commented. And I want him get into the -- more of the details. But one of the reasons we exported a bunch of ammonia in the first half was because even though our supply agreement with Mosaic was scheduled to begin in the beginning of the year, their vessel wasn't ready and you need a Jones Act registered vessel to be able to move a product intra the U.S. And so when that vessel becomes available, a lot of our excess ammonia will get soaked up by the contract. And so we won't really be in the position to need to export those funds. The urea situation, I think, it's really an opportunistic one, which is to the extent NOLA is trading at a discount to international marketplaces, it makes sense for us to go ahead and export because we can realize better netbacks. And on the UAN front, it's very situational, depending in terms of what our opportunities look like and optimize them to haul. So it's not that we have a stated strategy to export X or Y or not, it's really very opportunistic about maximizing netbacks.

That's exactly right. And the other component to that is we've spent several years working on this to put ourselves in the position to be able to execute against it. And so what you're seeing is a result of work by the various teams with relationships and agreements and destination facilities to be able to move those tons in an attractive way for the company. And so, I think that urea, for example, some of those were just spot deals, we recognized an anomaly in the market in May and early June, executed quickly against it. So 90,000 tons of our Q2 urea position went to Central and South America for urea. The UAN was more programmatic where we have some nice agreements in place, and we feel very comfortable with our ability to move that. A lot of that is inventory management as well. We produce a lot of UAN in Donaldsonville. And we can ship it in various modes, but during this quarter, like Tony said, it's netback-driven. And the netbacks were fairly attractive. So I like the program. I like what we've put together. We're a fairly small team. We can execute quickly, have good relationships and enjoy doing that part of our business.

All right. And just as a follow-up. Is there anything you've observed in the NOLA inland product premiums? Any compression in urea spreads or UAN spreads?

I think, Tony mentioned that there's been normal Q3 and we're in a normal Q3. We're asking our customers to store that product for several months and pay cash upfront. And so that does come at a discount, but we're fully expecting and we saw this year the premium for all the products are holding and expect that into next year as well.

Now, I would say, Ben, in that regard, we have seen some other producers in market drop their price well below kind of our price, which if you look at NOLA versus where they launched at, you'd say there was pretty significant compression. But in those instances, we just walked away, because we didn't think that represented the accurate value and pricing of nitrogen on a per unit basis. And so I feel very comfortable in terms of where our price realization is going to be against a number of other producers inland.

And our next question comes from the line of Chris Parkinson from Crédit Suisse.

Sorry guys, I think, I got cut off there earlier.

We just -- we enjoy your questions so much, Chris, we want to space them out. Every 10 minutes, you'll have to ask another one.

You haven't even heard my question yet.

Bert (inaudible) on this a little, but I just want to dive in just once a further. Can you just further comment on the domestic selling behavior, over the last quarter or really the last year, and just, I think, it's pretty clear that it's fairly unsustainable, especially given some Middle East imports. Can you just comment on your summer fill program? And I think another question alluded to this. And just kind of comment on your pricing and potentially deterring imports from the East Coast, specifically some -- I think, some postings in Cincinnati. And just how do you think that is going to be able to alleviate the blow from certain new facilities, mainly OCI coming online during the next few quarters?

Okay, Chris. Welcome back. So summer fill, we did launch it for last week for UAN and we see this a success. We targeted a certain amount of tonnage to be available during the quarter and into Q4. And we've achieved that. We've seen balanced purchasing from small, medium and large customers. We have not built much to our export book yet, so that potential store remains out there. But we feel comfortable with what we launched at NOLA and then the step-ups to the Midwest. And I think pricing was pretty good. We launched at a level that we thought was attractive for our domestic customers. And you're right. The East Coast is our lowest netback in our system and our Louisiana vessel that we supply over there, we move, I wouldn't say significant tonnage, but we moved tonnage over there as a new market for us that we didn't participate in several years ago. And we think with our logistic assets, we're now able to participate and plan to. The Cincinnati option has been really good for us where we're able to move a significant tonnage up in there and store it and move it out by truck and rail as well as Stolthaven, which we work with BNS. BNS has been very supportive of our ability to move by rail into the eastern markets. So probably our biggest growth market has been that eastern seaboard and into the mid-corner region of the east.

But Chris, you're absolutely right in terms of thinking about what -- the Russian production because we're not pricing off of what other people are doing. North America remains an import driven marketplace. And so pricing really is established based on what's the imported ton gets over here from. So as I said, when some of the other producers out there kind of want to blow their brains out, we just let them do it and because our focus is really on the Russian tons that are coming over. And that's how we calculate where pricing should realistically sit, not based on what other people are doing.

And we think we're attracted with price. Coast, Gulf Coast, West Coast, and so you're right, these new plants are coming online. We think, they have a place in the market. And we're trying to continually grow the options and advantages to us to be able to take care of our business, run our plants at a high rate and move the product ratably and maintain inventory and maintain a pricing structure that we believe is appropriate, reflects an appropriate value for the product to our end customer, which is the farmer.

And our next question comes from the line of Michael Piken from Cleveland Research.

I just wanted to discuss sort of how you think about the fill program in the future. It seems like NOLA prices are probably less relevant and as OCI kind of reaches closer to full capacity, do you expect to have sort of different pricing for each region? Or how do you sort of think about how you establish the fill program going forward?

Well, I'd say one thing and then ask Bert to get into the details, Michael. But we -- on our fill program, we do it on a deliberate basis today. So it is really -- it's not one-size-fits-all, it's kind of depending upon what the costs are to get to different areas. And as I've mentioned, it's not whether plant X or plant Y is up, because North America continues to be import dependent and it's based on really what's the cost of the Russian tons getting over here. And so that sets the East Coast price and those East Coast tanks and then it sort of cascades in from there.

I think, the fill program will have a future. Because again, especially the liquid product like UAN, you're logistically dependent to move incremental amounts on a continual basis. And so we have to have relationships and an understanding in place with our customers of where value lies to move this product, because we don't want to sit on it for 6 months. We can sit on it for several months and we can do different options with the plants in terms of product portfolio, but this is the business we're in. So we need to align with our customers and that's what we've done is sit down and try to figure that out. Regarding other producers, you have to remember, the one you mentioned is probably 100,000 tons a month. That's not that much in a 14 million-ton market. And so as a supplier, we believe, we're a global supplier and this is a global market that we're always looking at all different options and trying to maximize the profitability of this company.

Okay. Great. And then just another question regarding kind of the future of fall ammonia. I mean, how do you see that market with all the regulations evolving over the next few years? And is it better for CF longer-term to have a big fall ammonia market or is it better for it to be a more spring dependent UAN, maybe, having a bigger percentage of the market?

I would say, on the second question, is it better? We're ambivalent on that. We want to do what's right and we want our customers to do what's right for them and we do believe in the 4 Rs, right place, right time, right product, right rate. And fall ammonia plays a nice component in that. It's very good with adequate soil moisture applying at the right ambient temperature of 50 degrees, fall ammonia sets in and is available to the farmer in the spring. So it's a good utilization. I'm talking full utilization of his equipment, diesel, time and value of the product. It's going to be interesting this year in the fall because if you roll into November in the application period, today ammonia represents an exceptional value, multi, I'd say a decade-low value for the farmer. And so we should see if price and decision-making does have a correlation. Low prices should incent more people to apply, giving good weather conditions this fall, more ammonia. And by the way, we're just the company to take advantage of that. We're pretty excited about it.

We have the largest, as you know, Michael, ammonia distribution network between our terminals and our plants. And so a robust ammonia application season plays well against our asset base and we're well situated to take advantage of it.

But no matter what happens, we have a balanced portfolio for ammonia of industrial, our contract with the Mosaic, our ag business, our storage ability and then the ability to export, if needed.

Our next question comes from the line of Matthew Korn from Barclays.

So we talked a lot about imports and the effects thereof and it sounds as though you're seeing some effect of greater availability x CF within North America. My question is, does this get heavier before it gets easier, in other words, is there a good amount of domestic supply outside of your sales. It's new, it's been commissioned, but it's only now or into the second half starting to bite?

So what's come online so far has been our production. And other than that, nobody is fully operational that we know of. But we know Agrium came up with some new tons for urea in border. We think they're up in operating, the Waggeman Facility for industrial ammonia from Dyno, has been up for the last year. We think OCI is up and down. I don't think, they're fully operational yet. And then you have the Koch and the Dakota gas tons that are ammonia to urea. And so we believe those tons can be absorbed into this market. We're still going to have to import 3 million or 4 million tons of urea, 1 million to 1.5 million tons of UAN and some ammonia, no matter what. And so the import effects, I think the imports will continue to play a role, just probably less so. And that should give options for CF domestically as well as internationally and that we'll take advantage of.

I think, the bigger challenge a little bit, Matthew, is that we have a very robust and well-developed distribution network where we can store tons and kind of deal with the evolving buyer behavior. So when buyers are sitting on the sidelines, there's other things we can do, whether it's -- at first developed our export optionality or putting the tons into our storage system or whatnot. Most of those other new plants that are coming up don't have those options available to them. And so because they don't have a robust distribution network and they don't have export optionality, they're sort of stuck a little bit. And therefore, the thing that they tend to do is just drop price until they find some takers. And that does create a little bit of sloppiness. But again, we're not pricing off of what other people are doing. We're pricing off of the next ton into the market, which is, in the case of UAN, really a Russian ton.

Let me ask then about your own plants. With the current market requiring such a priority on liquidity, when you're spending $400 million in CapEx over the course of the year, are you delaying spending that you're accruing for '18, '19 and '20 at this point? Remember, the $400 million also includes about $25 million to $30 million on the Donaldsonville DEF project. So we've got 400,000 tons of DEF now on the urea equivalent basis at diesel, which gives us both great margin and product flexibility -- optionality. If we cut things down to just purely kind of maintenance capital, keep the lights on, we can tighten our belt a little bit off of that number and we're continuing to focus on that as we look forward. Clearly, minimizing our fixed charges is top of the list and Chris and his team on the manufacturing side are well aware of that and really focused on doing that, that exact thing.

And our next question comes from the line of Adam Samuelson from Goldman Sachs.

A lot of ground has been covered today. One thing that I didn't see in the slide deck that you have put in, fairly consistently recently, has been kind of the view of the cost curve. I'm wondering if you had any new thoughts or any updated view on where kind of marginal cost, I presume, Chinese anthracite will be sitting today and as you look into '18.

Yes. We don't really have anything that's dramatically different from what we've put out there in Q1. As I think most people know, we've added a new board member, John Eaves, who is the CEO of Arch Coal and one of the -- John has a tremendous background, but one of the things that's really terrific in the near-term is his understanding of coal markets globally and his visibility into what's going on in China. And I think that's kind of reinforced our position and we don't really have any changes from the Q1 material.

That's helpful. And then maybe a market structure question. And you think you alluded in your third quarter and second half outlook remarks that imports should not be coming here, given again where U.S. prices are versus the rest of the world and the increase in domestic production. And I'm wondering, as you look at that seaboard market globally x the U.S., if you think that there's enough liquidity there from key buyers in India, Brazil, Europe and others to really be able to absorb a reduced U.S. presence in the seaboard market. I'm just trying to think about how do we get U.S. prices back to international parity without cargoes finding the way back to NOLA because of liquidity that is often found there? Any thoughts there?

Yes. That's kind of spot on, right? So if there was circa 50 million traded tons last year and a slice of that came to the U.S. and we don't need that anymore, where did those tons go? And that's the nail on the head. And I think what happens is you end up with higher cost facilities in some of these other regions, for which the imports are an option, but historically haven't played a role. Those plants start coming under pressure. So in particular, YARA is shutting down a plant in France. There's some other challenges with some other higher cost kind of marginal European production, Ukraine is down. China has taken a big slice of production down. And so I think it's a combination of both less traded seaborne tons because the Chinese production is coming off. And also, new destinations for some of that. Instead of NOLA, it will go other places. But that's why we see some ongoing kind of sloppiness in the market likely persistent to the back end of this year and into next before we start getting a more robust pricing environment.

And our next question comes from the line of Steve Byrne from BofA.

Tony, you mentioned earlier what you view as important is the net capacity adds and as you define that metric, you have this roster of Chinese capacity, 2 million to 3 million tons you expect to be permanently shuttered over the next 2 to 3 years. How do you come to a landing on that roster? And how would you assess the potential upside if the -- none of that restarts versus the downside of all of it coming back on the stream. How do you -- do you have feet on the ground that are giving you that assessment and maybe if you could just highlight that a little bit?

Yes, so Steve, we do have some people on the ground. We have a Chinese national imports organization that does travel over there with some of our product managers. And we do have a page in here on Page 32 that gives at least a listing of some of the tons and where they are sitting and what their feedstock is. And so it's based on kind of an understanding. We did a pretty thorough analysis a year or 2 ago, plant by plant build-up of what feedstock was, what the cost was, where the coal or, where the feed came from and what the destination was. Was it being used internally or was it going basically as an export oriented plant? And it's a combination of some of that analysis along with where current coal prices are. And honestly, the restriction and challenges around gas availability and some of the environmental regs that have come up. And it's the source, the intersection of all of those that have led us to kind of this view of what capacity is kind of likely in jeopardy over there. Now, can things change? Sure. It could change either direction, but I think there is a concerted effort by the regime to transition away from some of the zombie industries to get some improved environmental policies put in place. And I think that plays very well relative to where -- what we're projecting and may even give us some upside.

And Bert, you're talking earlier about the shift in buying patterns in North America to more of a just-in-time delivery expectation, shifting to the spring, et cetera. Do you expect that trend to continue? And in general, do you think that could lead to perhaps just lower fill pricing versus historical levels? And perhaps, tighter market conditions in the spring?

We've seen these patterns ebb and flow over the years. As customers perceive value, they enter the market and they want to buy. And we used to have to limit the size of our fill program for UAN and at one point, it was 3 million tons and we had to limit it at that. And those were healthy prices. Last year, we almost couldn't find a buyer. And we didn't do almost half of what we've already done this year. And so each year is different. Again, as a perceived value, the carrying opportunity for that retailer to realize a higher price in the market once he sells. And so I think that's why we've just been cautious and we've been pragmatic through this process of trying to understand the market and utilize what we can. But I do believe that they're in the business of selling -- buying and selling and applying fertilizer for their farmer customers. And I think that will continue to grow. And I think we're moving towards a much more service-related side to that business than just a buying and storing. And we see those to our -- that moves to our advantage and moves of our -- those are the type of customers we want to deal with. So we see the fill program and the programs like that continuing to be an integral part of what we do.

The only thing I would add to that is, fill at $125 NOLA, for instance, is on a unit of nitrogen equivalent of about $180 urea and urea is $10 above that. So historically, UAN trades at a premium to urea when it gets upside down, then there tends to be a lot of interest. And we don't think that there's a lot of downward pressure coming on urea. We think, it's found a pretty reasonable place to trade at, absent a spurious extra vessel. But that would be a local regional issue, not a global one. And therefore, when UAN price drops to a place that it's very competitive or priced below urea, we would expect there to be a pretty high take rate on fill. So it's lower than it has been in the past, but we're still making pretty good cash margins at $125 and we feel comfortable with that.

And our next question comes from the line of Vincent Andrews from Morgan Stanley.

Just taking the Chinese urea production another level. Do you guys think it's possible that China would choose to become a net importer of urea? Or do you think they sort of would want to stop at self-sufficiency?

I think, it's highly possible. We were actually following that story and a little bit this year, I think as you look at the cost structure, and what Tony mentioned earlier, with the transition that's taking place in their economy, I think you have to take a little longer-term view. But longer-term could be a few years. There's adequate supply of low-cost urea coming out of the Arab Gulf that's logistically attractive and where China is based in terms of their production base is more to the south and interior, but your consumption base is in the north. And so there are places where it makes sense to import product on an actual cost basis. And so I think, as this rolls out and we see some of these shutdowns that we're expecting and not any new plants being built, there's a high possibility or probability that imports -- and it doesn't take much. 1 million tons here, 1 million tons there would be accretive to the market.

Okay. And to follow up to that if we take Slide 14 and just sort of get ourselves out to 2019, 2020, 2021, if we hold sort of current production cost constant, what do you think the U.S. cost advantage is on a short ton of urea basis versus the high cost of production and that's actually operating when we get out sort of once the new capacities coming on slower than normal.

I mean, part of that, Vincent, is embedded in kind of the evolution of LNG markets and gas markets and so forth, which we've spent certainly sometime kind of looking at. I would think in the neighborhood of $3.5 to $4 of gas spread is kind of a fairly reasonable sort of way to think about it. And the cost curve kind of continues and continues to evolve and will hold, but if you think about sort of roughly 25 MMBtu per metric ton of production, it takes more than that in the older plants. But in the current ones, the world scale ones, you're talking somewhere in the neighborhood of $100 of delta on cost before you basically load it on a ship and send it and then unload it. So that could be $120, $125 pretty easily by the time something gets landed over here.

And the next question comes from the line of John Roberts from UBS.

Back on Slide 32. I just want to confirm, you have none of the temporary idled capacity entering the permanent closure category by 2019?

That's correct. We don't have it in there. Again, it might be a bit of a conservative assumption based on sort of what's happening from a cash loss perspective, but that is the approach that we've taken on this one.

Do you have a guess at temporary idles for 2018? And does price have to get above the 2016 levels for anything to restart here, given most of those temporary idles occurred in '16?

So again, kind of our earlier comments, was we expect challenging market conditions through the balance of '17 and into '18. So if you impute what you will from that comment about pricing, but given -- if you got sort of status quo on coal-based cost, then I certainly would expect some of those temporary shutdowns to persist.

And our next question comes from the line of Sandy Klugman from Vertical Research.

A lot has been covered, but maybe you can help us think about the market potential for diesel exhaust fluid, discuss how the margins there compared to what you can generate in the agricultural market. And then if there's a sizable premium, how should we think about the potential for it to be competed away?

Sandy, we love to -- I'm gonna let Bert talk about the market size, but our price realization per urea equivalent ton is close to $100, between $80 and $100 depending upon where you are. One of the benefits that we have is because of the number of locations that we produce it at, we can provide supply security to our major customers. And if you're a one-off plant, you can't do that. There's still a fair number of, "suppliers of DEF" that are either bringing in prilled products and melting it down or transporting it in from -- transporting it in from Europe or other places, which is an expensive way to get it over here. So we operate, we think, on a pretty advantaged platform for a couple of perspectives in that regard. And it does take a fair bit of capital as we talked about at diesel in order to be able to produce it. So it's not free and it's not something that everyone can just run out and do. And even if they have it at 1 location, it doesn't mean they can get the same kind of supply agreements that we can given the surety of supply that a number of our key customers are looking for. So we feel we have a pretty sizable advantage in that regard. But Bert, you want to talk about the evolution of demand and how we expect that to happen?

It's been a very nice product. Everything you articulated is exactly where we are and we're excited about the growth. I would say we're at close to 1 million tons of urea equivalent tons of demand. And we expect that to continue to grow. The Class 8 trucks, the power units that drive this today are not even 50% converted and that will continue to grow. And then you've got the off-road rail, marine. If they do grow, I think that's a little more several years out. But for the first time, as I said in my remarks, that U.S. will pass or North America will pass Europe in total demand. But we expect both markets. And then if you get China to grow, Brazil, some of these other areas, longer-term, that's just positive growth and positive environmental benefit. And so we talk about the 2% growth rate in this business. A lot of it is industrial growth. People only focus on the ag, but, I would say half of the growth is now in the industrial segment and that will continue to be the case. And so the premium, we're on all the Class 1 railroads. We're able to move the product. We've got a huge fleet now committed to DEF, great relationships and contracts in place with our customers. So we have positioned ourselves well for this market.

The only thing I would say, Sandy, is 1 million tons today growing at a pretty good clip, is predicated on a dosage rate that's in the sort of 2% range -- 2%, 2.5%. If it gets up to 5% dosage, which it certainly could, we could easily see over 2 million tons of demand. So we think there's a fair bit of runway and that's one of the reasons why we built such a large capacity of diesel from a DEF perspective.

That's helpful. Just a follow-up question on your gas cost in Western Canada. Does the company have long-term projection for where the AECO spread to NYMEX will sit?

I don't think we have a long-term projection. We like it. I mean, it's trading at times at a $1 below the hub and when you look -- as you march your way up through the Midwest and into Canada, each of those steps are positive or I'd say lower gas cost for those plants. And with the gas production capable of coming out of Canada as well as the Bakken, we anticipate that AECO is going to stay at a fairly attractive rate to Henry Hub.

Yes, I think, Tony, one of the things that we look at is as the infrastructure gets built out in the Marcellus Basin, we're able to move more product up into the East Coast from that basin, that pushes gas from Canada, from the east back West. And that's just sort of accentuates that issue. Obviously, you're going to have more gas coming out of Permian and some other places, but on the States, so gas that had been coming to the U.S., given everything that's happening here, from Canada, it's going to get backed up. And as that happens, we would anticipate that AECO spread will remain robust.

And our next question comes from the line of Jeff Zekauskas of JPMorgan.

I think, you generated about $20 million in cash, exclusive of tax benefit that you got this quarter. Now, maybe, your payables are lower than they might be and maybe your inventories are a little higher than they might be. As a base case in the second half, do you think you're going to use working capital? Or are you going to throw off cash from working capital?

Jeff, this is Dennis. If you look at what I said with respect to the third quarter, fourth quarter, I don't know where we're going to be, but it's not impossible to believe that we may be using cash in the second half. I suspect we will use cash certainly in the third quarter. But with respect to the second half as a whole, it's hard to say because it's really so dependent on what happens to price. If you look at last year, we had a price rally in the fourth quarter going into first quarter of this year, whether that materialize this year or not and what the magnitude of that might be, is going to really be what dictates a lot of that for the half.

The other point, Jeff, that I remind people of this, we generated about $150 million of cash in the first quarter from operations. And then another $20 million in the second. So kind of year-to-date, we're up -- not that far away from $200 million of cash. And so if we dip into a little bit of that here in the third quarter and then see what develops in the fourth, we feel pretty comfortable about that.

Yes, but I think the important point, Jeff, is even with these challenging conditions of the cash, we are generating and covering our fixed charges. We just have a huge liquidity position as we've talked about.

So many of the fertilizer companies have been under all kinds of pressure with the nitrogen downcycle. Do you think that these sorts of pressures increase the probability of further consolidation in nitrogen fertilizer or do you think that they just create so many price inefficiencies for the valuations of companies that consolidation opportunities are diminished in an environment like this.

I mean, I think consolidation is likely to occur. It was pretty public that one of the public fertilizer companies was evaluating strategic alternatives and looking for sale that they canceled that process. But I think that that's not uncommon. If there's basically $1 of synergy to be had in any given situation, $1 synergy goes a lot farther in a lower premium environment, lower premium because the base assets are priced lower to begin with than it does in a high valuation environment. So I would expect there to be ongoing consolidation.

Ladies and gentlemen, that is all the time we have for questions for today. I would like to turn the call back over to Martin Jarosick for closing remarks.

Thanks, Daniel. And that concludes our second quarter call. If you have any additional questions, please feel free to contact us.

Ladies and gentlemen, thank you for participating in today's conference. This does conclude the program and you may all disconnect. Everyone, have a great day.