Good day, ladies and gentlemen, and welcome to the 9-Month and Third Quarter 2019 CF Industries Holdings Earnings Conference Call. My name is Kevin, and I'll be your coordinator for today. [Operator Instructions]. I would now like to turn the presentation over to your host for today, Mr. Martin Jarosick, CF Investor Relations. Sir, please proceed.

Thank you, Kevin, and good morning. Thanks for joining the CF Industries 9-Month and Third Quarter Earnings Conference Call. I'm Martin Jarosick, Vice President Investor Relations for CF. With me today are Tony Will, CEO; Chris Bohn, CFO; and Bert Frost, Senior Vice President of Sales, Market Development and Supply Chain. CF Industries reported its 9-month and third quarter 2019 results yesterday afternoon. On this call, we'll review the CF industry's 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 Tony Will, our President and CEO.

Thanks, Martin, and good morning, everyone. Last night, we posted our financial results for the first nine months of 2019, in which we generated adjusted EBITDA of $1.3 billion, after taking into account the items detailed in our earnings release. These results reflect higher year-over-year global nitrogen prices, lower natural gas costs and continued strong execution across all elements of our business. I especially want to highlight the great work of the CF team. Throughout this year, we have run our assets well, managed through logistical challenges and ensured our customers received product when and where they needed it. Most importantly, we did it all safely. Our 12-month rolling recordable incident rate was 0.61 incidents for 200,000 work hours, significantly better than industry averages. Our team's great work, combined with positive industry fundamentals, drove a 21% increase in adjusted EBITDA compared to this point last year, and we continue to efficiently convert the EBITDA we generate into available free cash flow. On a trailing 12-month basis, our free cash flow is $830 million, which today provides our investors with an industry-best free cash flow yield of 8%. As we've said before, we believe we will generate superior free cash flow through the cycle compared to most of our global competitors. That's because our cash generation capability is built on an enduring set of structural and operational advantages. Our structural advantages are clear: access to low-cost North American natural gas, operating in the import-dependent North American region, the largest and best production and distribution network in North America and the long-term demand growth for nitrogen. We also consistently focused on increasing our operational advantages to drive further margin growth. We do this by investing in our assets and people, and actively managing the company through the cycle. The cumulative effect of…

Thanks, Tony. The CF team continued to perform at a high level during the third quarter, positioning us well for the remainder of the year and spring 2020. First and foremost, we met customer needs in July as the spring application season in North America was drawn out due to poor weather earlier in the year. Following the extended spring, we focused on building a good order book. This included a well-received UAN fill program, which launched about a month later than normal and was met with strong demand. Demand for other products was positive as well, and we essentially shipped what we produced in the third quarter. As a result, we ended the quarter at seasonally low inventory levels. This gives us flexibility in the months ahead as the fall and spring application seasons develop. Looking at spring, we continue to anticipate strong corn plantings in the United States. The current soybean-to-corn futures ratio supports higher U.S. corn plantings in 2020 and is comparable to last year's ratio at this time. We are ready for the fall application ammonia season to begin which has already started in some areas. We believe our customers are expecting a positive fall ammonia season, given expected strong corn acres and attractive nutrient pricing. As always, the weather will drive how positive the fall season will be. If a good application window opens, we believe farmers will take advantage of it. If a good window does not open, we have ample storage capacity to position product to meet customer needs in the spring. We expect that the remarkable stability we saw in the global nitrogen prices this year will continue into 2020. Global demand has been healthy overall and has required additional tons for marginal producers in China to be bid into the market.…

Thanks, Bert. In the first nine months of 2019, the company reported net earnings attributable to common stockholders of $438 million or $1.97 per diluted share. Our EBITDA and adjusted EBITDA were both approximately $1.3 billion. Our trailing 12-month net cash provided by operating activities was approximately $1.5 billion and free cash flow was $830 million. Cash and cash equivalents on the balance sheet at the end of the quarter were over $1 billion. Since the beginning of the year, cash on the balance sheet has increased by $337 million after investing $297 million in sustaining capital expenditures, repurchasing about 5.7 million shares for approximately $250 million, issuing $200 million in dividend payments and distributing $186 million to noncontrolling interest. Given our liquidity position, which as of yesterday was approximately $1.2 billion in cash and cash equivalents, our strong cash generation and positive outlook into 2020, we announced earlier this month that we will redeem the remaining $500 million in senior notes due in May 2020. Additionally, last night, we announced that we will redeem $250 million of our 2021 senior secured notes in December. Retiring this debt is the latest step and the balanced approach we have taken over the last 2 years to manage the company, prudently allocate capital and return to investment grade. These actions will produce -- reduce our gross debt by $1.85 billion. They also support our focus on reducing fixed charges in order to provide us the greatest long-term capital flexibility through the cycle. As you can see on Slide 10, our annualized fixed charges in 2020 will be $186 million lower than they would have been without the steps we have taken since 2017. This includes reducing annualized interest payments by about $121 million, which achieves our goal of annual interest payments below $200 million. It also includes lower level of dividend payments due to share repurchases as well as the elimination of the cash distribution to Terra Nitrogen unitholders. Additionally, we have focused our capital expenditures on safety and reliability. Not only has this supported our industry-leading asset utilization rates, but has also kept our capital expenditures for the past few years at around $400 million per year. These actions, along with the industry recovery from the trough conditions of 2016 and '17 have greatly improved their credit metrics. As a result, we believe we have built a strong case to earn investment-grade ratings. With that, Tony will provide some closing remarks before we open the call to Q&A.

Thanks, Chris. Before we move on to your questions, I want to provide some summary comments to frame how we're thinking about the future. We had a great first 3 quarters of 2019, with adjusted EBITDA increasing 21% year-on-year. On a last 12-month basis, we generated $830 million of free cash flow, which is truly a fantastic year. We also have the highest conversion efficiency of EBITDA to free cash flow in the industry. With all the free cash that we generated in the last 12 months, we have returned approximately $750 million to shareholders through share repurchases and will have retired an additional $750 million of debt by the end of this year. That accomplishes our objective of bringing our balance sheet back into investment-grade metrics and drops our annual interest expense well below $200 million per year going forward. Looking ahead, we're excited about our outlook for 2020. While there are always moving pieces, we think that overall 2020 will be similar to 2019. As we've explained, we expect our sales volumes to be similar from year-to-year. Our 2020 global cost curve projects average nutrient prices to be in a similar range as 2019. And the NYMEX Henry Hub forward curve suggested natural gas costs in 2020 will be lower, which all means, we expect another fantastic year in 2020. And since we have already repaired our balance sheet, all of the free cash flow we generate will be available for us to deploy for value-creating growth or to return to shareholders. This company is a highly efficient cash-generating machine. We've driven over 10% accretion for our shareholders in the last 2 years by investing in accretive growth and share repurchases, while fixing our balance sheet at the same time. With our balance sheet now investment grade, we look forward to driving additional shareholder value by continuing to invest in accretive growth and further share repurchases. With that, operator, we will now open the call to your questions.

[Operator Instructions]. Our first question comes from Adam Samuelson with Goldman Sachs.

So I guess, first, Tony, Bert, I'd be interested just to get some perspective just on the near-term kind of market dynamics as we think about the fall. I mean, you've seen urea prices fall somewhat counter seasonally in the last month or so and NOLA started to trade at a pretty healthy discount to some offshore destinations and just get your views on the drivers of that and kind of what would get that -- get that back to parity? And then I just have a follow-up after under this -- on the decision to repay the debt, if we could.

So when you look at the market today, as we look at the 9-month performance of CF and the industry overall, we've seen a fairly stable market, operating in that $240 to $260 short ton NOLA range and on a metric ton probably the same $240 to $270 FOB year of Gulf-type range. And so coming off of those numbers in the last month or so has been a bit of a surprise, but I do think this goes to -- it's a global market, and we've bid in some tons due to some of the India tenders and some of the changes. You've seen the India number -- volume purchased increase but as well seeing the Chinese export numbers increase. And so we would have predicted a 2 million to 2.5 million ton export coming out of China. It has turned out to probably will be at 4 million to 4.5 million ton export, so a doubling coming out of China. Why is that happening? Looking at the cost of production there with energy cost being coal, it's down about 8%. And then you look at the RMB, the devaluation, averaging probably around 7.10 now. So it's another 5%. It has allowed the Chinese producers and some of the higher cost producers to sell at around $250 a metric ton FOB. So those tons have made it into the market, and you've had low gas costs in Europe through the summer and through Q3. So combined, you probably had higher operating rates globally, which has pushed the exportable ton, there's not a finite but kind of an average of 45 million metric tons that are traded annually, a few more tons into that market mix. And so we have traded below international parity recently, and we think that's an anomaly. We're probably significantly below where -- we've been averaging, let's say, $10 over the last year, and that number today at $210 to $215 a short ton is probably $20 to $30 below the international market. We see increased demand coming or continued demand coming from India, another tender or two, so another 1.5 million tons demanded there. Brazil is behind on their imports, another 1 million or more tons needed in Brazil. And then we think Europe will step in, and we're looking at a very positive 2020 with 93 million to 94 million maybe even 95 million acres of corn, so that incremental demand -- and I'm trying to figure out what will happen with the ammonia season that could go to upgraded products and push then urea higher. So I don't think this market lasts -- could last through this quarter, but that's -- our position is that we're able to bridge over that. We have inventory space available. We think there's a lot of buying left to be done, and then we'll see what happens in the spring.

Our next question comes from Christopher Parkinson with Crédit Suisse.

Given the magnitude of your cash flow generation. Can you just remind us of 4 capital allocation priorities outside of share buybacks, including any potential for low-risk, high-return brownfield as well as the potential for M&A? Just on the latter, are there also -- there have been -- there's been some news of a few potential assets floating around in the U.S. as well as Europe. So can you just comment on your willingness to do something a bit larger as well as your general stance on asset bases outside of the U.S.?

Yes, Chris. I mean, look, I think from a capital allocation perspective, we have said for a long time we want to get the balance sheet back into investment-grade metrics. And with the recent announcement, I think we're there. We've got managed our annual fixed charges down to a level that it's very comfortable, even in sort of trough conditions of '16, '17, we'd still be net pretty significantly cash-flow-positive during those kind of trough conditions. And we think that, that's representative of an investment-grade rating, and we're very comfortable with the sustainability of the balance sheet through kind of down cycles. So once that was behind us, then we look at, obviously, sustaining capital to maintain what is the highest operating rates in the industry on our ammonia plants, is this number one call for capital; number two would be, if we've got accretive growth where we can buy assets in a way that we believe creates value for our shareholders, and that would be number two. And in the absence of doing those things, then I think we look to return cash to shareholders in the form of share repurchases as our preferred mechanism, given that we already have a pretty robust dividend that's in place today. So that's kind of our priorities. We're relatively open to geographic expansion. I do think when you start clustering assets together, you're able to better realize synergies from larger network effects than if you've got them spread around. But I think at the end of the day, it all comes down to as long as there's a set of assets that we believe we can run well and really leverage our organizational capabilities against, then we'd be open to considering a number of places. But it all really depends upon price point. And they've got to be at a place where we feel it's creating value for shareholders because our alternative is to buy back more of what's already the best asset base in nitrogen of the world, which is our own share. So instead of overpaying for poor quality assets, we just buy more of our own.

Our next question comes from P.J. Juvekar with Citi.

Just quickly, Bert, I think you talked about increased China exports and falling coal prices, do you think that puts a lid on urea prices in the range of whatever you talked about $250 per metric ton. And then a question on your cost curve. Your cost curve is delivered prices to the U.S. Is that the right way to look at the urea market in your mind? Because most of the Chinese exports go to India. So maybe I can get your thoughts on that?

So regarding increased China and is that a lid, this is a constantly changing and evolving discussion. If you go back several years to the peak of exports where they were dominating 35%, 38% of the global trade. That was a significant impact, and we saw the impact of that driving NOLA prices down to $150 a short ton in the United States. And so when you look at some of the changes that are taking place in China, I think it's a positively evolving issue. The recent devaluations due to all these trade conflicts or our questions, is that sustainable? Is that desirable? I don't know. I would say they would probably be better off trading in a range where they were in the 6 80s [ph]. But when you look at what's happening in China with capacity coming off the peak capacities, where today our estimate is probably 80 million tons of capacity, there's estimates as low as 70 million. And then the operating rates has hit 70% per the publications. We think they're operating in the 65% to low-60 range. So that makes available today, probably 52 million to 53 million metric tons. And then when you look at what ag is fairly constant. Industrial, also fairly constant, and it's the export swing volume. So we don't see the capability to move a significant amount of tons. And we say many times in some of our investor discussions. If you idled these assets over a period of time and don't idle them appropriately, it's very difficult and very expensive to bring those assets back into production. And I don't think that's an attractive thing to do basically export energy from China and continuing to pollute both water and air. So I think that China is stable at that range of 2 million to 4 million tons. And I think, economically, it is not attractive today to export at these levels. And yes, on the cost curve, your second question, we do look at it because we're primarily North American producer and participant. And so that's the market we really want to focus on with 90-plus percent of our tons staying in North America. But you're right, those Chinese tons are basically staying in Asia. They are trade restricted now to North America at their peak, probably 1.5 million tons came to North America. That's not happening, it won't happen. So I think that -- again, as India has been fairly consistent at 6 million to 8 million tons, they're probably going to draw 1 million to 2 million to 3 million tons from China if the Iranian sanctions go away, maybe some from Iran and the rest from the ARM Gold for North Africa. And then we see the world market balancing, and that's still 45 million metric tons of export demand, and that does drive the global cost curve.

Our next question comes from Vincent Andrews with Morgan Stanley.

Just wondering if you have any sort of latest thoughts on the potential for the Indian capacity to come back. There have been some moving parts on some of those facilities going forward, some of them not? And then there appears to be more lower cost gas available to them. So just your latest and greatest thoughts there would be helpful?

Yes, just as the previous question, that's also an evolving issue in our industry and with Prime Minister Modi's decision to be India-centric and India-driven and industrial policy revamping -- and it's not a revamp. We visited, and we've been following these plant additions, and we'll be there next year. These are new constructions, basically. Now you have plans for today that are new and idled. The Matix plant is still not running. And there are pipelines to be built and infrastructure to support these plans. So will they come on. We have them in our S&D going forward. And I think that will be a hit to imports, probably going down a couple of million tons. But the long-term projections today, natural gas globally is inexpensive. The North American shale producers have done a good job of driving that down globally, which is positively impacted Europe and some in Asia, specific to China. And I think India will be a net beneficiary in the short term. Longer term, that's not going to be the case. And so it gets back to the cost curve. And as the Indian government willing to subsidize these plants to an extraordinary amount or not, and we believe they won't. And so I think that this import demand will continue to be in a range, probably, let's say, 5 million to 8 million tons, but that's in 2 or 3 more years. So more to come.

Our next question comes from Ben Isaacson with Scotiabank.

On one hand, you talk about new supply not keeping up with demand growth; and on the other hand, when we look at coal prices in China, they've gone from, I think $12.80, a couple of years ago to $11.80 at the start of the year, now they're at $10.80. Can you talk about how you see coal prices developing? Why have they come down? And does the government really manage this within a range, how low can they actually go?

If you look at coal costs, I mean, that's another globally traded commodity, and that's driven by several things: one, the cost of production; two, the cost to move the product; and three, the cost to receive the product. And so the downturn in Chinese costs both domestically -- but the portion that they are bringing in from Australia and any other place that they could, we don't believe can stay as low as it is. And so we don't see a further decrease much broader from the range we're operating in today. The other issue, driving costs up, are the changes to bunker fuels and fuel costs and shipping, which we see positively going -- positively for us going up. And then I think over time, it's the issue of what is the right energy source for a global economy that's growing. And as coal the right source, there is still additional plants going in. But longer term, we believe that, that would be less, and there's been announcement that China is trying to increase their gas production through their shale opportunities. And we would say that, that would probably be a fuel source of the midterm future. So we'll see how that goes.

I mean, I think the other thing I would add to that is, with LNG prices having fallen dramatically on an MMBtu equivalency, coal price has to come down in order to be viable. And China, at the end of the day is still a coal-driven economy. And so back to Bert's point, while there is, I think, a bunch of external factors that are -- have provided some pressure on coal, there is increasing demand for coal utilization. And we think, ultimately, that ends up supporting prices there. So we're not terribly concerned about the top end of the cost curve, doing something dramatic.

Our next question comes from Duffy Fischer with Barclays.

Just a question around demand in North America. So starting in the fourth quarter last year, we all talked about the bad fall application season, we talked about the short window in the spring. Given the crops we planted this year in North America, how much do you think we shorted the North American market on nitrogen? And then if we get the bump to, let's say, your $94 million midpoint acres of corn next year, how much more does that add? So kind of 2 buckets of incremental nitrogen growth over the next 12 months versus the last 12 months, if you could help break those 2 buckets out?

So looking at North American demand, it's a use it or lose it type. We say nitrogen is the only nutrient that is absolutely necessary, and you can't carry it through from one season to the next. So you're right, last year, Q4 was not a good ammonia season, and we've had a couple of those in a row, and we needed to make that up in the spring. When we moved in the spring of 2019, it was wet. It was difficult, and ammonia did get down at some form and fashion, but a lot of that moved to upgrade, urea and UAN. So the total in applied, I don't have that number in front of me. But I don't think we shorted it. What I do believe is that we exited Q2 with low inventory throughout the system, both retail and producer and imported inventory. And so when we're looking today and out the window in Chicago, it's snowing. It's not conducive to ammonia application today, at least in our backyard, but we've generally applied ammonia through early to mid-December in most years, and we would look for a warming trend. And the temperatures look conducive in Iowa and Nebraska and Southern Illinois to get that season done. But as I said in my prepared remarks, if that eventuality is -- we're unable to do that, we believe that at least at CF, we have the ability to make it through into spring, and that makes that even much more of a challenge. And then that adds value to our distribution and end market production assets and our logistical ability to move product quickly into the market, and I know there's going to be a timing game. So that's what I would say, we -- the corn-to-bean ratio is attractive for this 94 million acres of corn. And what's going on in the protein market, I think that's going to continue to be attractive.

Our next question comes from Mark Connelly with Stephens.

Good morning. This is Joan Tong on for Mark. Just question -- quick question on nitrogen application practice, in general. It seems like nitrogen application is a key area of focus in digital ag as well as some of the ag tech trend. Are you seeing any of those new technology or newer products affecting the way farmers are buying or applying nitrogen or affecting their interest in pre-buying?

When you look at these new technologies and the new biostimulants or biologicals or different things, everything is in the testing phase, and it's in the theoretical and they're starting to put out what that could be in the practical. And so for us, no, we have not seen changes in application. What we are seeing, and this is over years with precision ag; with the education around when to apply; and the 4Rs [ph], which is the right place, the right time, the right product at the right rate you're seeing very good farming practices. And there's a whole education going on in our industry, which we are funding and supporting. And that is also focused on watershed improvements and the issues of just good environmental practices. And that's being, I think, spread amongst companies amongst groups and especially with custom applicators, such that you don't see fall application of ammonia starting until about right now when the soil temperatures are at the right -- the optimal temperature and moisture profile to hold that nitrogen. And so we looked at these positive changes. And if they were to achieve what they're promoting, these new technologies, we see that as accompanying nitrogen consumption and then you'll have a yield boost, which I think is good for the farmer and good for the low-cost North American producer or farmer being able to compete in the global economy.

Our next question comes from John Roberts with UBS.

Back to the earlier question on capital allocation. New plants typically have some low-cost debottleneck opportunities. When you get to the first major downtimes coming up for your new plants, do you think we'll see some capacity expansions kind of on the order of magnitude of 10% or more?

John, no, I think the issue right now is based on operating rates globally of ammonia plants and where Tampa ammonia price has been settling lately, we look at ammonia debottlenecks as not a terribly interesting proposition from a standpoint of a return on investment. Now that said, there is a significant uplift in margin per nutrient ton going from ammonia into the upgraded products, urea or UAN, and we are looking at potentially debottlenecking on the upgrade side. But that -- the benefit of that is it doesn't add any nutrient ton capacity to the global S&D, it just changes the form of nitrogen. And those kind of debottlenecks are, I would say, relatively efficient from an overall cost perspective and largely fit within our $400 million to $415 million a year. It wouldn't take us outside of that band. So I think those are certainly things that we're looking at and considering, but it doesn't really change the cash flow dynamic on the business or what we've talked about.

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

In the Democratic primary, there's been a bit of a switch of the frontrunners. And one of the frontrunners that could win here has expressed the concern about fracking. We could see a change in fracking in the states in the next little while if couple of things work out that way. Have you started to think about what your risk appetite maybe for gas hedging? Or how you have to change your strategy if it's a change in that kind of political stance? And would you want to get ahead of some of that things?

Yes. I mean, I think there's a real question as to whether or not the authority that's be espoused by that individual actually exists within that office or whether that requires -- that's a -- I think at some level, that's more of a Supreme Court-related issue, because it's more of a state's rights versus federal government thing. So it's not obvious to us that, that power actually exists within that office, but that's for other people, I think, to decide. And I'd say, that is a huge kind of macro U.S. economy kind of decision if it was made. It wouldn't affect not only our business but the economy as a whole. And I don't expect -- I think it's a lot of rhetoric, actually. I don't actually expect something like that to show up.

Our next question comes from Steve Bryne with Bank of America.

Tony, you were just talking a little bit about nutrient tons shifting. I just wanted to ask a little bit more about -- if you look at your gross margins by product, it looks pretty slim in ammonia. Do you have the ability to shift more out of ammonia? Or at this time of year, you need to move some ammonia just because it's a product that you can move now? And also related to ammonia, just wanted to ask you about the impact of the Magellan pipeline closure, does that have maybe a differential impact on your competitors in terms of distribution costs, more so than it does for you? And does this lead to a higher corn belt ammonia price longer term?

Yes, Steve. I mean, I think a couple of things are going on. I would focus more on the 9-month numbers for ammonia than I would on the third quarter because the third quarter really has virtually no agricultural ammonia. And so all of the ammonia that moved in the quarter were really driven off of industrial kind of contracts, which tend to be much more Tampa-based pricing, in general. And the Tampa price has been pretty low, which why -- which is why we're not that excited about further ammonia debottleneck because they just -- they don't pay out. We do run our upgrade plans kind of at 100% capacity. And so that's why if we're going to be able to shift more ammonia and to upgrade, we're going to have to be doing some of the debottlenecks that were asked about earlier, I think, by John. But we have a great end market distribution network for ammonia. And you saw that in the second quarter when we moved a lot of that volume and got very good price realization and very good ton movement. Clearly, I think what you've seen is with the Magellan pipeline going down, you've had coke and Enid announced a big urea debottleneck expansion, which again is reducing the amount of excess ammonia that they have. You've seen other people make moves away from anhydrous into upgraded products. We have the benefit and verdigris of being able to barge ammonia out of that plant, either for use elsewhere in the system or to be able to export it. And so I think we still have a fair bit of flexibility relative to the plants and other people are making moves to try to reduce their dependence and cost structure of moving ammonia around the system. But our end market terminal is to provide a really nice lift for us during the application season, but the spring tends to be the big application. As Bert mentioned earlier, as we look out the window here in Chicago, it's snowing, so I'm not sure we're going to see much of a fall application season this year. But what you don't get in the fall in the way of ammonia just put increased value on the upgraded tons as we get into spring.

And the only thing I would add, Steve, this is Chris, is that as you look at the ammonia segment, as Tony mentioned, looking at the 9 months is for -- probably more indicative, but if you look at the adjusted gross margin, you'll see that really a large part of that was depreciation, and it's pretty much in line with the prior year quarter. Additionally, the tons are up a little bit. As Tony mentioned, that Q3 is a higher maintenance period. So we had slightly higher ammonia output than we would have in a typical quarter.

Our next question comes from Jonas Oxgaard with Bernstein.

From what I can tell at current ammonia prices, your Trinidad plant is running at negative margins, and that looks like it's going through into your equity line as well. Is there room for renegotiating the raw material pricing? Or how you're thinking about that in the current environment?

Yes. I think, for the most part, Jonas, the Trinidad plant that you're seeing at the negative margin is related to a tax amnesty program that we, along with our joint venture down there in that partnership, agreed which was related to withholding taxes from the year 2011 through current, and that was about a $16.5 million settlement that related to our portion of that. I think when you look at the Trinidad asset, it's still producing at a cash margin, given that the cash cost of natural gas down there is related to ammonia as well.

And we're actually one of the few people that have -- look, we announced, I think, about a year ago that our 5-year extension with NGC, National Gas Company of Trinidad had been extended. And so the price at which we're buying is commensurate with historical kind of Carib-based gas deals. Most of the other producers on the island have annualized through their contracts and have had to renegotiate. So we're -- we believe we're in a favorable cost position relative to the other producers on the island. And as Chris said, I think the results you're seeing there and that's a -- is a onetime tax issue as opposed to an ongoing operating performance issue.

Our next question comes from Michael Piken with Cleveland Research.

I wanted to talk a little bit more about the UAN situation. I know you guys had a good build program, but it seems like efforts to increase prices have been a little bit challenging going forward. How are you sort of thinking about the fill program kind of going -- or not the fill program, but UAN sales going forward, in terms of the timing and what we might expect in fourth quarter versus next year?

Yes. So as I said in my prepared remarks, we were pleased and continued to be pleased with the fill program because of spring planting being late and wet we were -- we carried applications well into late July and inventory was fairly low when we launched the program in the last week of July, first week in August. And we built a healthy book which allows us to operate the plants full and utilize our distribution assets, railcars, trucks, barges as well as vessels because we're still participating in the international market. And so as we look into, I would say, today, we're looking into Q1, I think we're very well positioned. As I mentioned, our inventory has continued to stay low. We have low gas costs. And UAN has been ranged bound in NOLA. It has traded close to what the publications are talking about, maybe on a little bit on the high side. But as we look towards spring, we believe that the interior -- and it's already reflecting this a larger margin spread in the interior. So we're constructively positive, UAN and the demand. And again, just as in last year, if the ammonia is unable to go down, it makes it very challenging to get that much in onto the ground. And so we're planning to participate in that market. Now we've done some things. You're correct that the UAN, the sanctions in EU, which we've been communicating about. So you can see from our numbers, we've cut our exports almost in half, the majority of that being decreased participation in Belgium and France and the EU markets. And we've repurposed a lot of those tons to different places, and we've been expanding our distribution reach in the United States. And so we think we're prepared for the future. And if that future is heavily reliant on North America, we think we will do that and do that well, but continuing to participate in a small way or smaller way in the international market.

Our next question comes from Don Carson with Susquehanna Financial.

Just a question on your price outlook for next year. You talked, Bert, about how you thought prices might be flat, but was that a NOLA comment because there's some very strong pricing in the corn belt this year due to all the river issues? Are you expecting a repeat of that in 2020? And then just quickly, are you taking advantage of some of these higher offshore netbacks in urea by increasing your exports out of diesel currently?

So looking at the price outlook, we were fairly benign in our comments saying that we're projecting 2020 to look fairly close to 2019. And that's due to the review of the cost curve and what we think is capable and where tons will move and the puts and takes of the business. So yes, it was more of a NOLA comment. River close has already taken place. River close means that barges can no longer traverse up to the northern territories, and if they do, they are stuck up there until spring and they have to pay storage. So where we are? Again, we think inventories are low, and a lot of P&K is in inventory space, which makes it a little bit difficult for urea to get into the dry spaces. So we think that will be a positive outcome for us. We have a lot of space available to our -- to ship to in our pine bend and Medicine Hat in Port Neal, production and storage facilities. But when you look at the previous question on exports, we're off about 50% from our exports from comparing 2018 to 2019, and the values just weren't there relative to what we could achieve in North America. If these values continued to stay where they currently are over the last month, then we would participate more in the export market. We're ambivalent to where our tons go, it's margin-driven. So if we can get $1 or $4, $5 more overseas, we'll load it up, and we can ship a lot of product out of Donaldsonville.

And Don, the question about the end market premium, which did kind of blow out during the spring, given logistical challenges and how we're viewing that kind of next year. I think, given the end market distribution and production that we have, we always expect to at least capture kind of the transportation spread but what we find typically during high demand periods or short replenishment windows is that that's when the power of that network really shines. And the last couple of years have seen more volatility as opposed to more smoothness in terms of the operations and what's required and demanded and what the market is willing to pay for. And I think that there's a little bit of more of the same kind of the assist embedded in terms of how the spring runs, particularly when right now you've got NOLA trading at such a discount to the international space. What that really means is the very few imports are coming this direction because if you're an exporter out in the Middle East or someplace else, you can get better values by going to other parts of the world. So North America is not paying enough to attract imported tons at this point, which means that, that's going to further sort of stress the system relative to be able to resupply in the market, and again, that's when our network shines. So from our perspective, the fact that during the low demand volume quarter, Q3 prices lag a little bit relative to the international markets. Net-net, that's sort of not a bad thing.

Our next question comes from Brandon Dempster with Consumer Edge Research.

I just wanted to talk about maybe what you think is the range or the price point rather in North America that you'd have to hit or maybe sustain that would bid in greenfield capacity? And maybe that's thinking more in a 3- to 5-year horizon?

Well, I just -- I just don't see it happening. And again, anyone that's interested in building one, come -- send them our way, and we'll sell them a plant at replacement cost. I'm happy to do that. I think you got to get urea prices sustainably up north of $350 on average for a full year to even think about it. And then if you're going to do that, you're still better off building in Nigeria or Russia than you are building over here just because labor costs are so uncontrollable here.

And I think, just to build on Tony's comment, that's a sustainable price above $350 in order to get something that's in low teens type of return for what they would be spending on labor and equipment.

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

Thanks everyone for joining us, and we look forward to seeing you at the upcoming conferences.

Ladies and gentlemen, this does conclude today's presentation. You may now disconnect, and have a wonderful day.