CF工業控股 (CF) 2017 Q3 法說會逐字稿

Good day, ladies and gentlemen, and welcome to the Third Quarter 2017 CF Industries Holdings Earnings Conference Call. My name is Vince. I would be your coordinator for today. (Operator Instructions) I'd 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 the CF Industries Third Quarter Earnings Conference Call. I'm Martin Jarosick, Vice President, Investor Relations for CF. With me today are Tony Will, our CEO; Dennis Kelleher, our CFO; Bert Frost, Senior Vice President of Sales, Market Development and Supply Chain; and Chris Bohn, Senior Vice President of Manufacturing and Distribution.

CF Industries reported its third quarter 2017 results yesterday afternoon as did Terra Nitrogen Company L.P. On this call, we'll review the CF Industries' results in detail, discuss our outlook and then host a question-and-answer session.

Statements made on this call and in the presentation on our website that are not historical facts are forward-looking statements. These statements are not guarantees of future performance and involve risks, uncertainties and assumptions that are difficult to predict. Therefore, actual outcomes and results may differ materially from what is expressed or implied in any statements.

More detailed information about factors that may affect our performance may be found in our filings with the SEC, which are available on our website. Also, you will find reconciliations between GAAP and non-GAAP measures in the press release and presentation posted on our website.

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

Thanks, Martin, and good morning, everyone. Last night, we posted our financial results for the third quarter of 2017 in which we generated adjusted EBITDA of $134 million after taking into account the items detailed in our earnings release. I am particularly pleased with our performance and results for the quarter, especially considering how the quarter began with prices meaningfully below 2016 levels.

As nitrogen prices rose from the unsustainable lows of the summer, our team responded extremely well, dramatically ramping up shipments and capturing the improving prices. Our team demonstrated impressive execution across all parts of the business. Operationally, our plants ran really well once again. We produced 2.5 million tons of gross ammonia, but most importantly, we did it safely. Our 12-month recordable incident rate was down to 0.85 incidents per 200,000 work hours. Our team has also achieved more than 3 million hours without a lost time injury. This is a tremendous achievement considering the level of activity across our system.

Our sales volume was the second highest quarterly volume ever. This is particularly impressive because the third quarter is historically a lower-volume quarter for us due to the seasonality of applications in North America. During the quarter, we participated in the North American in-region demand, sparingly at first, but with increased enthusiasm as prices rose. This included our successful UAN summer fill program.

We also leveraged Donaldsonville's ability to export to a greater extent than ever before as North American nitrogen prices remained below global parity. During the third quarter, we exported 1 million product tons, which is a new CF record for any quarter. These export sales allowed us to balance our system and manage inventory positions to maximize overall margins. When North American customers were unwilling to purchase products at netbacks comparable to international prices, we increased exports and built inventory by pulling back from selling in North America.

As prices began to recover, we chose to sell more actively and were able to capture better pricing on very strong volume. Our export capability, combined with our extensive inventory and distribution network, are unique to CF and are critical elements to maximizing cash generation in this business.

We have also additional ability to grow export volumes even further should North America continue to trade below international parity.

As I have mentioned, global nitrogen prices strengthened significantly during the quarter from the unsustainable lows of July. Producers in high-cost regions curtailed production, and global supply was also reduced by planned and unplanned outages in a variety of regions. When demand increased during the quarter, especially in Brazil and India, product was in limited supply and global prices rose. Prices in North America rose as well but remained below international parity, and global trade flows rapidly adjusted to this pricing discount.

During the third quarter, far less imported product arrived in North America than in the first half of this year or the same period in 2016. In fact, North America was a net exporter of urea in July and approximately net neutral on imports/exports for the third quarter as a whole. The significant price fluctuations we have experienced in 2017 are symptoms of a market in transition as new capacity ramps up and global trade flows adjust, and this transition is not yet complete. This is particularly true in North America, where we expect the final set of capacity additions to ramp up to full rates by mid-2018. Should importers not fully take this additional capacity into account, we could end up retesting the nitrogen pricing lows we saw in the spring and summer of 2017. However, after we get through 2018, we continue to expect global demand growth to exceed new capacity growth and set the stage for a more sustained nitrogen price recovery.

We believe our performance this quarter demonstrates how CF is the best-positioned company to benefit from the recovery. Our operational advantages, absolute scale, production flexibility, significant in-region storage and export optionality allow us to set global volume -- to set volume records and maximize cash margins in a period of seasonally low demand in North America.

Our structural advantages, access to low-cost North American natural gas, operating in the import-dependent North American region and the long-term demand growth for nitrogen are well established and enduring. Taken together, we believe CF will benefit disproportionately from the improving price environment we see ahead of us.

With that, let me turn the call over to Bert, who will talk more about the commercial environment during the quarter and our forward view, then Dennis will discuss our financial results before I offer some closing remarks. Bert?

Thanks, Tony. CF's performance in the third quarter demonstrated the strength and agility of our team and system. We shifted our product mix when needed, employed our logistics and storage assets and carefully selected opportunities of when and where to sell. As a result, we maximized the overall cash margin available to us while achieving close to record shipments. We believe we are positioned very well going into spring 2018.

The quarter began with NOLA urea trading at approximately $155 to $160 per short ton. These price levels were the lowest seen in a decade, excluding 2009, and were at or below full production cost for many producers around the world. Prices remained low through July, with the expectation that the quarter would close at similar pricing levels. However, the continued low level of Chinese production and exports, combined with a weaker U.S. dollar and energy and production cost increases in various areas of the world, helped to lift prices through August and then move higher in September. As we exited the quarter, urea prices around the globe continued to show strength, supported by low inventories and higher-than-expected demand in India and Brazil.

However, North America lagged the rest of the world in terms of pricing and demand throughout the quarter.

While urea prices followed the sharp upward trajectory, ammonia and UAN did not increase as dramatically. CF again achieved very solid export volumes for UAN and ammonia and a record level for urea. Late in the second quarter and early third quarter, the netback available from international destinations increased and exceeded the netbacks available for domestic shipments. The team was able to pivot to exports and build a book for the various products that allowed CF to load and ship record volumes at favorable values.

We exported 200,000 tons of urea, 300,000 tons of ammonia and almost 0.5 million tons of UAN. These shipments allowed us to improve our results in 3 ways. First, we delayed the launch of our UAN fill program, so we could capture higher prices for June and July applications. Second, we did not participate in the extremely low ammonia prices initially offered by some for fall applications. And third, we were patient with urea sales and were able to capture higher values later in the quarter.

Additionally, Mosaic began taking ammonia shipments from Donaldsonville early in September at the contract price, which was favorable to prices available for either domestic or export destinations. Going forward, the amount of ammonia we expect to have available for export will decline significantly due to the supply agreement with Mosaic.

Now let me provide some thoughts on our broader outlook. The global market remains in a period of transition as new capacity comes online and global trade flows adjust. These changes make it harder for participants to interpret market signals, and this leads to greater volatility. Buyers in North America remain risk-averse with 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 changes play to the strength of our North American production and distribution network. They also make import options less attractive as internationally-sourced material often involves longer lead times and importers bearing significant inventory risk.

In any case, we will continue to work with our long-time customers as well as our growing list of new customers to meet these challenges and deliver tons to the right areas at the right time, utilizing storage and adjusting product mix. We have demonstrated our ability to do just that and are well positioned as the global nitrogen market continues to transition.

With that, let me turn the call over to Dennis.

Thanks, Bert. The company reported a diluted loss per share of $0.37 and EBITDA of $139 million for the third quarter. After taking into account the items detailed in our press release, our adjusted diluted loss per share for the third quarter was $0.39 and adjusted EBITDA was $134 million. Our adjustments included a $7 million unrealized net mark-to-market gain on natural gas derivatives.

I want to highlight the cost reductions we have achieved across our system. This is a result of our targeted efforts to reduce costs in our head office and in our plants. Our SG&A was roughly flat compared to the same period last year, demonstrating the scalability of our business model. We increased our capacity by about 25% and sales volume by 33% since last year without incurring significant incremental head office costs. In fact, despite adding significant production, we have among the lowest SG&A as a percentage of sales in the fertilizer space and commodity chemicals generally, as you can see on Slide 12.

Looking at SG&A on a per ton basis, we reduced SG&A per product ton by 23% since 2015. At our manufacturing sites, reduced plant level activity and better procurement practices have decreased professional services, payroll and maintenance costs, resulting in a lower controllable cost of sales.

As you can see on Slide 11, we have reduced controllable cost of sales per product ton by almost $12 compared to third quarter 2016 and more than $15 on a year-to-date basis. Our balance sheet remained strong. And our cash and cash equivalents are $2 billion as of September 30, and our $750 million revolving credit facility was undrawn.

Monday, we announced that we will redeem the May 2018 bonds on December 1, 2017. We estimate that the total cash payment will be approximately $817 million. This action fulfills the commitment we made over a year ago to redeem these bonds with cash and reduces our annual cash interest expense by $55 million.

Capital expenditures for the third quarter of 2017 were $105 million. For the full year, we expect to spend approximately $375 million for new activities in 2017. As of -- additionally, as of September 30, 2017, approximately $158 million remained accrued but unpaid related to the capacity expansion activities in 2016. Most of this unpaid amount originated from disputes with certain contractors and vendors.

On our second quarter earning call, we provided our thoughts concerning the third quarter and our view that it was unlikely that we would exceed prior year's adjusted EBITDA. This was based on our view that the benefits of higher volumes and lower controllable costs would be more than offset by higher gas costs and lower product prices, and that July sales volume, somewhat driven by the poor pricing environment, were fairly low.

Compared to that starting point, prices, particularly for urea, improved dramatically in August and September, as Bert described. Additionally, our team took advantage of rising prices and offshore demand to sell and ship approximately 300,000 product tons more than we expected. We also achieved greater cost reductions than we expected. As a result, our adjusted EBITDA for the third -- for the quarter came in substantially above our expectations at the time of the second quarter call.

I will now provide some specific thoughts regarding our expectations for the fourth quarter of 2017, understanding that this, too, is subject to changes in market conditions. We expect to see higher volumes and lower controllable costs per ton compared to the fourth quarter of 2016. We also expect higher nitrogen prices across all products and for gas cost per MMBtu to be essentially flat compared to the same period last year. However, I'd like to point out a few things worth mentioning.

First, quarterly sales volumes are difficult to predict, and we believe it is better to think in terms of 6- or 12-month time frames. It may turn out that some of the robust volumes we achieved in the third quarter pulled forward volumes that we otherwise would have sold in the fourth quarter. Second, some of the volume we sold in October was priced before the price runup in the third quarter had fully materialized. And lastly, the fall ammonia season is just beginning in much of the Corn Belt. A grower's decision to apply ammonia is very sensitive to weather conditions during a narrow window of time. Therefore, a significant amount of volume and price risk for ammonia remains in the fourth quarter.

So based on our view of all of these factors today, we believe it is likely that our financial results for the fourth quarter of 2017 will exceed the $133 million of adjusted EBITDA reported in the fourth quarter of 2016.

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

Thanks, Dennis. Before we open the call to questions, I want to highlight a couple of things. First, our business has demonstrated its resiliency even during the most difficult environment our industry has seen in over a decade.

Slide 13 shows how we've grown cash on the balance sheet by $828 million through the first 3 quarters of the year. Even excluding the federal tax refund we received in the second quarter, along with a one-time voluntary contribution to our pension program, we increased net cash by $72 million. This means that cash we generated, even during this challenging period, has more than covered our fixed charges, including CapEx, interest, dividends and distributions to noncontrolling interests. With the bond repayment Dennis mentioned, we will further reduce the interest portion of fixed charges by $55 million per year.

The second item I want to highlight has been our team's execution over the last 15 months. On Slide 26, we compare the achievements discussed today to the commitments we made on our August 2016 conference call. At that time, we laid out 4 targets for the business: first, complete and start up the expansion projects by the end of 2016; second, reduce annual CapEx to less than $450 million in 2017; third, reduce our controllable expenses; and fourth, reduce leverage on the balance sheet by repaying $800 million of debt due in May 2018. Our team's efforts have yielded excellent results on all 4 targets. We completed the expansion projects by the end of 2016, and all new units are reliably running more than 15% above nameplate capacity.

New CapEx in 2017 is expected to be approximately $375 million. We have reduced controllable expenses per ton by 16% and SG&A per ton by approximately 23% year-to-date.

Finally, earlier this week, we announced that we will redeem the 2018 bonds early, reducing our leverage. Importantly, we achieved all this with excellent safety performance. Given the heightened level of production and shipping activity over this period, along with normal turnarounds and maintenance, that is an outstanding accomplishment.

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

(Operator Instructions) Our first question is from Adam Samuelson of Goldman Sachs.

Maybe first for Tony. Just some thoughts on the market and, really, product pricing and premiums as you've seen them evolve over the third quarter and, really, this year, where the U.S. has continued to trade at a sustainable discount to the rest of the world; the UAN since the summer has traded at a discount to urea, which is fairly unusual. Can you talk about kind of how you think those evolve? What, in particular for the UAN premium, gets that reversed? And how to think about your own realizations against that construct with -- your realizations versus benchmark are actually quite good this quarter.

Sure. Thanks, Adam. This is Bert. And the market has evolved in a little bit of a strange way. Coming off the lows, as we mentioned, of urea and looking at that point in July when we were at $155, $160 and launching our UAN fill program at the end of July to where we are today is remarkable. The market is up close to $100 in North America and internationally, even more.

And you're right, we're trading at a discount to the world, as frustrating as that is. But I do think that's a reflection of this transitionary period that we've been discussing, where we have a little bit of a stop-and-go type market with purchases and positioning. And so we've become adept at accessing all markets at all times and in conversations with customers around the world and working with different customer partners to move our product at the margins that we deem appropriate.

And so when you look forward for product pricing, we're positive to what's taking place around the world. There's still capacity to come up in North America, and there's idled capacity in different parts of the world that will probably come up, off of turnarounds and maintenance, in late in Q4, Q1. But we've seen robust demand to support that, and this current situation in India, with the tender that has just closed and been priced, we think is attractive. You still have a few more months of Brazilian activity as they prepare for a second crop, corn planting in February and March, so imports will arrive probably into early February there. And then you have the North America season starting at that point, with a significant amount of imports still to take place to support planting in spring.

Relative to the UAN summer fill program, we are very pleased. The team did a great job of working with our customers, broadening our customer base, our geographic imprint. But you're right, UAN has traded at a discount to urea, and that is historically an anomaly, and we do see that over time and believe that UAN is, we believe, a very good product. It will return to its premium status, but that may take a while as we talk about this transition year in 2018.

The only thing I wanted to just add that a point that Bert makes often on the UAN side, which is below -- urea is kind of ubiquitously traded. UAN is principally only used in a few markets, so you've got North America, a couple of markets in Europe and Argentina, and we're just beginning in Brazil and a few other places. But there's much less international demand for it.

And with the new Donaldsonville plant operating at over 6,000 tons a day, that's over 2 million tons of capacity we brought on. Wever has brought on more than 1 million tons. And because at Port Neal, we were urea-limited in terms of our UAN, with the new urea plant we've brought on, we've added a couple of hundred thousand tons of UAN capacity in Iowa as well. So when you look at that the U.S. has added kind of 3 million, 3.5 million tons of capacity, that's a huge percentage increase, and I think it's just going to take a little while for the marketplace to absorb that before you see the urea-UAN kind of pricing premium renormalize.

And that is the benefit of, as we talked about, our product agility or product mix configuration. We can produce all that Tony talked about or move some of that to dry urea, liquid urea, DEF and we're balancing utilizing those different tools.

Our next question is from Don Carson of Susquehanna.

Bert, a question on just the domestic market. I mean, to what extent was some of this price rise we saw in Q3 kind of a scramble by dealers to source product as they saw that Enid and Wever weren't going to fully ramp? And what's your view on the premium between NOLA and the Corn Belt on urea, UAN and ammonia, as we go forward? I mean, should your strong exports maintain that premium? Or do you think it comes down with all the new capacity in-market?

So looking at the premium today, I'll answer that part first. It's holding and we're positive to the premium. There's a direct cost to move product, no matter if you're moving it from Galveston or NOLA, or the East and West Coast to hit the consumption markets from, let's say, Ohio to Nebraska, the Dakotas to Texas. And we're fully intent on that premium holding because, again, there is a cost.

Now ammonia has had some lower prices than historical in terms of the spreads, in terms of the absolute product values as a divisor based on their nutrient content. I think that's an anomaly also. I think as we balance this, a few of the producers probably had excess ammonia going into the quarter because the upgrades weren't prepared and needed to move that product. So I think this fall is an anomaly based on what will probably happen in the future with ammonia applications for corn.

But you're right, we've seen a nice move in the market. I do think several of our big customers as well as medium and small customers did position themselves well at the UAN fill. And -- but we look at -- as urea, we've had fewer imports and probably fewer produced tons than anticipated, so I do think there is a market that needs to be fulfilled going into Q2 to be ready for spring. So I do think there will be some needs, a need to be purchasing as well as logistically moving those tons over the next, let's say, 4 to 5 months. So that's why we're constructively positive.

Let me just add a couple of thoughts to that, which is, the U.S. still is going to need to import another what, Bert, about 3 million tons of urea or so?

I would say, 4 million.

4 million tons. So you can't get there if North Africa is trading at a significant premium to where we're trading. The prices at NOLA eventually have to rise.

The other point that I'd make up is we've demonstrated, this last quarter, our ability to export Donaldsonville. So in a lot of ways, Donaldsonville tons should be considered really an international plant because we're not going to just pay the transportation cost and move those tons into the Midwest and not recover that cost. We could move those tons into the international marketplace and get better netback.

And so that's another reason why, from our perspective, the whole question about whether the Midwest premium holds is a bit of an odd one because we're not going to incur that cost when we've got other options, and the Midwest is chronically short products even if we put all of the D'ville stuff up there. So someone's got to pay the freight to move the product up there. We see no economic reason why there's even a question about the Midwest premium.

Your next question is from Christopher Parkinson of Crédit Suisse.

You guys have clearly done a pretty solid job reducing controllable costs. But can you comment on whether this is mainly focused D'ville and Port Neal after their respective ramps, any startup costs, et cetera? Or was it fairly broad-based across your facilities in the Southern Plains and even Canada? In other words, just how much better do you believe you can get in 2018 and 2019?

This is Chris Bohn. To answer the second part of your first is that, as Tony and Dennis mentioned, these initiatives we've put in place last year in the third quarter, so we're beginning to see some lapping of it. So I would expect that we'll still see some benefits, specifically from volume since we didn't bring on Port Neal until late Q4 last year, so there will be some volume benefits that we continue to see.

As far as being focused just at the new expansion projects, I think you're correct in the assumption that this is broad-based. What we really did is we started to focus and be much more diligent on our spending, primarily across professional services, engineering activity and maintenance and payroll expense. And we've seen that come down at all our facilities, so it's not just directed at those 2.

Chris, one other, just a bit of follow-on there. So we've got a fair bit of engineering capability at the plants. Historically, we've had a lot of that focused on de-bottlenecking and improvement on the capacity side. As we've looked at a more difficult market environment where the value of the incremental ton is perhaps not as high as it used to be, Chris Bohn has done a great job of redirecting that talent base away from improvement, which then is -- begets more CapEx and expense on the professional service side to really focus on maintenance and turnaround activities, so that further reduces our expenses going through those activities. So it's I'd say a double benefit that we've been able to achieve there.

Our next question is from Joel Jackson of BMO Capital Markets.

Can you talk a little about -- so ammonia sales were quite strong, you've obviously had a lot of capacity. I think you spoke generally a little about possibly some pull-forward volume for some products in Q4 and Q3. Say, for ammonia, was this for pull-forward ammonia, maybe give a little update on that dynamic?

No, it wasn't. The ammonia sales in Q3, Damian did a great job of, as we talked about, this whole export capability and leveraging, where we talk about moving 300,000 tons in absence of the Mosaic contract which we had planned on. Their vessel was late, so we had to pivot and do different things, and one of those was working with some of the destination markets for those sales.

And so we do have some ag sales of ammonia in Q3, and the majority of that is for weed applications in the Texas, Oklahoma, Kansas area. And so pulling forward ag sales in Q3 is pretty difficult because the corn application start -- have already started by late October and will run through, hopefully, into December.

And so we're anticipating a normal ammonia season still to be determined. We're looking out the window today in sunny Chicago to rain and clouds and although we're not in corn country, we are following the weather maps and it's colder up in the northern territory like the Dakotas, but for great application in Nebraska and Iowa. So we're anticipating the ammonia season to kick off and be strong for the next 30 days.

Our next question is from Steve Byrne of Bank of America.

This is Ben Gottesdiener on for Steve. Can you discuss your estimates of channel inventory levels of nitrogen in the major grain-growing regions of the world, U.S., Brazil, India and how that compares to historical levels?

Sure. When you're looking at India, we had projected, based on 2016's performance of tenders, there was an absence of tenders from September all the way to April and very limited import volume. That then pushed excess inventory into the market in the beginning of 2017.

What we believe happened was a significant drawdown of inventory in India, and you're seeing that represented today from September through December, a tender a month or more, and additional consumption of probably 1.5 million to 2 million tons just in India. So we think that, that inventory, and it's represented by the statistics, is low in India and they will continue to tender probably even into January, positive for the market.

Brazil is up about 1 million tons to date. Difficulty to store urea in Brazil over time. Imports started early in March and April and has continued forward to exceed 4 million tons for this year. So we're positive on Brazil, but don't see them carrying a lot of inventory past February. And then looking at our market, we believe coming out of second quarter in June entering in July, inventory for all products was very low. And hence, we had the positive pull for our UAN fill program, at we thought appropriate pricing level around $125 of FOB NOLA.

And so we're shipping against that, but we've had an incredible amount of exports and probably limited production as reported by some of the publications in some of the newer plants that were supposed to come online. So I would expect that inventory is at an average to below average level for UAN and urea. For ammonia, that mostly stays with the producers, and our inventory is adequate and we're prepared for a robust fall, should that happen.

Our next question is from P.J. Juvekar of Citi.

Corn prices have gone up in China even before the winter. So what's your outlook there? And then, secondly, on China. If -- China has shut down significant amount of anthracite capacity. If urea prices stay about $2.50, let's say, how much of that capacity in China do you think can start back up?

P.J., relative to Chinese coal, we're pretty constructive on where that's headed. One of the folks that we've added to our board recently is John Eaves, who's the Chairman or CEO of Arch Coal. They obviously watch that very, very carefully. We coordinate a bit of intel with them, and we've remained very constructive on Chinese coal prices, going forward.

I think the other issue is I don't think anymore it's just purely a matter of what the energy cost is in China. We really do believe that the current administration is very focused on environmental reform, and there are a set of those plants that just don't meet emissions requirements that won't come back regardless of where urea prices hit on the international front. Bert or Dennis, just...

I think looking at the coal cost on the structural basis for competitiveness, you've seen probably a $20 to $30 cost increase directly reflected in urea. That, combined with low production capacity utilization, 59, 60, 61, whatever the publications our industry has put out, puts production around 60 million to 61 million metric tons; exports, we project are 4 million to 5 million, leaves a domestic market of around 56.

And we believe there also, just like India, there was significant destocking that took place in the spring. And today, in several areas of China, the interior price is greater than the exportable price. And so I think $2.50 is a little low. I would say prices need to go probably over 300 for them to be bringing back any production. And I agree with Tony that, that production that's environmentally challenged, I don't know if the price is what brings it back to the market.

Yes. I think the other thing I would add is that from an exchange rate perspective, as we modeled into our cost structure, which you see on this slide, we're sort of at about 6.80 per dollar. If you go back to sort of this time last year, people were predicting it will be well above 7. And so not only have coal prices moved in the right direction, the environment stuff that Bert talks about, but in addition to that, the exchange rate has helped as well.

And our next question is from Vincent Andrews from Morgan Stanley.

On the export program, a couple of things. One, how fully baked are you in terms of the various regions or partners that you'd like to trade with? Are there more relationship you're still looking to build? Or do you think you're where you're going to be? And then secondly, is there going to be, going forward, a base level of volume that you're going to need to export regardless of the comparable netback, whether it's in the U.S. or in the export market, just to sort of maintain those export relationships? Or how nimble can you be?

I think we can be incredibly nimble. Based on the product mix that's available to the company and the spreads between ammonia, urea, UAN, nitric acid, ammonium nitrate, DEF, UL, I think those optionality as well as domestic options and inventory, when you take the whole matrix and look at what is available for us to move, do and work with, I don't think we have to do anything.

And yes, we are always building relationships, both domestically and internationally. And our customer -- I think, our total number of customers has doubled over the last 5 years. And that's on purpose. We've added sales staff, we've added regions, we've -- as well as added international staff, that we're actively participating and leveraging those relationships for the long term. We've worked with several traders as well as destination markets in Chile, Colombia, Brazil, Argentina, Uruguay and actively want to grow in that area because we believe in the long-term structural advantages of farming down there.

In the European market, Belgium, France, Spain, Germany also active for us on leveraging our U.K. operations with the people that we have based in [it]. And so when you look at that across the board, we say this all the time, but we participate in a global market. And anytime that there's an arbitrage or an advantage, we want to participate and know about it. And so we don't have a specific tonnage or allocation per market. It's all driven by netback and optionality.

But Vincent, to that point, I mean, back to Bert's point on netback and optionality. I mean, I think it's pretty easy for us to get north of 5 million product tons a year if the opportunity were better for us to export than it was to keep those tons home. I think year in and year out, it's probably going to be closer to 0.5 million tons per quarter.

But I want to be clear, we're not going to just export in order to make room for other people. I think people import product into this marketplace at their own peril because we're going to run our assets very hard, and you get on the wrong side of one of those trades and you could end up losing a lot of money. So we'd prefer our tons stay in here, and people need to be cognizant of that.

Our next question is from Andrew Wong of RBC Capital Markets.

I was just wondering, could you help clarify or maybe just provide some more details on the fourth quarter guidance? I mean, I think being above last year's EBITDA is pretty understandable with the new volumes and higher prices and maybe somewhat lower costs. But could you help me maybe on the magnitude or help quantify how much higher you see that? And then, just on the cost curve, how do you see that evolving out to like 2019, if -- assuming input prices don't really change things?

So Andrew, I'll tell you, we're just kind of dipping our toe in the water, relative to providing some thoughts around expectation. We don't go out more than sort of one quarter because as was evidenced in our second quarter call with respect to what we were thinking third quarter might end up being, we didn't see the magnitude of the pricing change, and that has a material impact on what ends up our results being. So I think we've said about all we want to say in terms of directionally where we expect to be.

As you get out to '19, '20, '21, there's an awful lot of things that start influencing what the shape of that curve looks like and the absolute magnitude of it. One of the main ones, as Dennis indicated earlier, is exchange rate, coal price, gas price in Europe, and so forth. So we try to give our best view of expectation for '18. But again, there's enough variables involved, particularly as you think about the possibility for tax reform in the U.S. and how that might affect strengthening or weakening U.S. dollars, how interest is treated in terms of deductibility or not, economic expansion.

All those things drive exchange rates, and that's a big input to that cost curve. So we haven't published anything that goes out to '21, but I would say, we feel very good about North American natural gas costs and we believe we're going to continue to operate deep in the low part of the cost curve, and we feel really good about where we sit.

Our next question is from Michael Piken of Cleveland Research.

Just wanted to touch a little bit more on the cost side here. If you could just give us a flavor for any updates on natural gas hedges in 2018, that'd be helpful.

On gas, we are hedged, as we exited Q3. As we've given previous guidance, we are 10% hedged for 2018 and we have our industrial contracts that are tied or gas-based for an additional 20%. So 30%, roughly, of our production is covered, and we have not added any additional hedges for Q4 or 2018 or forward.

Our next question is from Ben Isaacson of Scotiabank.

It's Oliver, on for Ben. Over the past couple of years, we've seen your AN and other prices performing pretty well relative to your urea and UAN. How much of an impact of changes such as GrowHow and sales mix into products like DEF had? Or is this mainly due to AN being a lot less elastic than UAN and urea? Do you see other drivers of the move?

I mean, I would say a couple of things -- and Bert, jump in. One of which is it's been a tremendous benefit for us to be the 100% owner of the former GrowHow business, the U.K., CF Fertilisers U.K. So we've brought a lot of discipline, both from an operational perspective and a marketing perspective, to that business.

And then separately, the beginning of 2017 was when the kind of long-term supply agreement with Orica/Nelson Brothers kicked in, and that contained in it some pricing provisions that were more attractive than historical. So it's the combination of those 2 things that have really made the AN business improve pretty substantially for us, relative to what it would have looked like a couple years ago and we feel very good about that segment.

Our next question is from Alexandre Falcao of HSBC.

I have 2 questions. One is, you've been painting a very positive scenario for 2018. My question is what would be -- if anything goes wrong, where do you think we're going to see the biggest probability of going wrong? And second, specifically on Brazil, seems to be a very strong market and you guys are still not fully penetrating there. Is there any -- it's just a ramp-up phase? Is there anything in particular, could expect more of that market from you guys, going forward?

So what could go wrong? Actually, I'd say what already went wrong in leading from '15 into '16, '16 into '17, were a strong dollar, low-priced, low-valued soft commodity markets, corn, sugar, wheat, cotton, low-vessel freights and a flattening energy curve and low-priced energy in Europe and most places in the world, cheap coal. What's changed? All those.

So as we go into the end of this year, freights are up. You could ship from China to the United States for less than $10. Today it's probably over $20, maybe $25. We still have a flat energy curve, but looking positive with coal pricing increases in Australia, China and I think a positive market there. A weakening dollar with the euro at one point approaching 1.19. And RMB also stronger, ruble stronger, those are good. Brazil, a little weaker, which is good for agricultural production down there.

And so what could go wrong is we've talked about, and this is why we believe it's still a transition year with additional capacity in North America coming up. Capacity was off-line or in maintenance coming up and demand patterns, going forward. So I think we're in the year, but as we talk about growth, as this market grows both industrially and agriculturally, growth eventually catches up and surpasses because there's not a lot of new capacity coming on, post '18. And that's when we believe the market becomes more of a demand-driven market than a supply-driven market.

Relative to your question on Brazil, Brazil is a great market. We have a lot of good friends down there, meet with them regularly. Matter of fact, one of our staff is there today. And we integrate and are active in understanding the demand patterns and price points. The reason we're not active in Brazil are 2 reasons. It's one of the lowest-priced markets in the world. And today, it's recovered. It's at $300, let's say, plus or minus metric ton.

But as you work back to freight, that hasn't been that attractive for urea and it's predominantly a urea and a sulfate market. We are trying to grow the UAN market and are consistently shipping product in there. So I don't see us ramping up necessarily to a market to be a participant unless it makes economic sense. And so when that changes or when that becomes attractive, you'll see us actively participate in a greater degree.

The other comment I would make about '18 is that I don't know that we've necessarily tried to paint an overly rosy picture about it. We've pretty much consistently said our view of '18 is going to be a lot like '17. There's going to be pricing volatility, there's going to be ups and downs as the new capacity that comes online is absorbed into the marketplace. And that leads directly into what could go wrong.

So you get a couple of traders and importers that think they're smarter than everyone else, they bring too much product in here, prices crash because they don't appreciate how much real capacity there is in North America, and that could certainly affect it. So until those kind of trade flows work their way through the system, there's going to be some uncertainty and volatility out there, and we're not counting on '18 being much different than '17.

Our next question is from Sandy Klugman of Vertical Research Partners.

Does the company have any thoughts on China potentially moving its gasoline market to an E10 blend? By my estimates it could translate out to 2 billion bushels of incremental corn demand, which would obviously be very positive for CF.

Sandy, we agree that, that would be incredibly positive for CF, and we are following that through our ethanol friends. There are exports taking place, as you've seen over the last couple years when sugar prices were high in Brazil and more sugar was committed to that rather than ethanol in Brazil, we were exporting down there, and I like how we're trying to build markets. Ethanol is a great oxygenate and a good product for some of these markets that are trying to make the transition, either away from ethanol or doing different things with their supply mix on gasoline.

So our take is, yes, that is positive but it's a long way to go. Whether it's a domestically-produced product, which again pulls more domestic-produced corn or imports, I think we're a couple years away from seeing that take place. It's probably E1 right now, and so more to come.

Our next question is from John Roberts of UBS.

When you say you expect continued price volatility, do you think similar lows and similar highs to 2017? And I ask that, primarily on the low side, because once you go below the cost curve, it's really hard to tell how low, low is.

Yes. I mean, I think, John, obviously we don't have a crystal ball in this regard. I think the lows kind of get tested when you have excess material floating around because you're in the shoulder periods between applications and you've got big slugs of new capacity that comes online. And then people have these loss positions that they're trying to unwind and get away from, and it really just drives pricing low.

But as you saw this year, prices can't sustain those levels for very long. People shut down. As that material gets absorbed, there's a pretty quick response back to more in the $300 per [met], which is kind of the global price point right now based on India tenders and what's being sold out of North Africa. So we do think volatility will be there, whether it overshoots as much as happened on the downside this coming season, I think a lot of that is going to depend upon whether importers and traders find some discipline.

Ladies and gentlemen, this is all the time we have for questions for today. I'd like to turn the call back to Mr. Jarosick for closing remarks.

Thanks, everyone, for joining us on the call. We look forward to speaking with you and -- at our next call in February.

Ladies and gentlemen, thank you for your participation in today's conference. This concludes the program. You may now disconnect. Everyone, have a great day.