Good morning, welcome to Yara's third quarter results presentation. Our presentation today will be by our CEO, Svein Tore Holsether, and our CFO, Torgeir Kvidal. After the presentation, we will have a Q&A, where we'll also be joined by our Head of Market Intelligence, Dag Tore Mo. Then I would like to have the pleasure of introducing Yara's CEO, Svein Tore Holsether.
Thank you, Tor, good morning to all of you. As usual, I want to begin with safety, as this is Yara's number one priority and an area where we have to improve further. Our recordable injury rate has dropped to half of what it was at the beginning of last year. However, in the first half of October, we have suffered two tragic fatalities in our operations. On Wednesday, October 11, Adrian Gutierrez Chinchilla fell from height of nine meters while working at our installation in Veracruz in Mexico. On Friday, October 13, Ailton Oliveira de Araujo, he was run over by a truck at our Salitre facility in Brazil. Our thoughts and prayers are with Adrian and Ailton's families, friends, and colleagues who have suffered these tragic losses. We are investigating the two fatalities with the highest priority.
I visited the Veracruz site last Friday, and I will travel to Salitre tomorrow evening. We cannot and will not accept fatalities or serious injuries at Yara, and these tragic accidents make us even more determined to drive and enforce our Safe by Choice way of working. As you can see on our slide, our TRI rate has dropped both in 2016 and in 2017. When you experience a fatal or serious accident, this improvement is completely overshadowed. Moving to the summary of the third quarter. Both ammonia and finished fertilizer production increased. Ammonia production was up by 7% compared to a year ago while finished fertilizer production increased by 8%. The Yara Improvement Program is ahead of schedule and has already delivered $ 210 million of the targeted $500 million of annual earnings improvement within 2020.
We're making good progress in all main parts of the program, and I will come back to this later in the presentation. Yara delivered weaker third quarter results, mainly reflecting lower commodity fertilizer margins. Earnings this quarter were in particular impacted by higher energy cost and a weaker US dollar. Yara's underlying earnings per share were 5% lower than last year. As already mentioned, the decline mainly reflects lower commodity fertilizer margins. Our reported earnings include NOK 139 million foreign exchange gain and NOK 343 million of negative special items, mainly the provision for closure of the Pardies site in France. Last year's reported earnings included NOK 114 million of foreign exchange loss and NOK 37 million of positive special items. Already mentioned, we had a strong production performance both for ammonia and finished products.
The 7% higher ammonia production is despite the Porsgrunn plant not producing most of the quarter following the fire in April. Finished fertilizer production increased 8% compared to last year, mainly reflecting reliability improvements for urea and nitrates, and also the expansion project in the Uusikaupunki NPK plant in Finland. We have posted several new monthly production records during the quarter, both at plant level, calcium nitrate production in Porsgrunn, Norway, and urea production in Belle Plaine in Canada. At Yara level for calcium nitrate and NPK. We also have record finished fertilizer production company-wide for the last 12 months. The Yara Improvement Program has so far delivered $210 million of cumulative annual benefits. Well ahead of our 2017 target of $150 million. The $210 million are measured at 2015 margins, and the equivalent number using year-to-date 2017 margins is $150 million.
We're ahead of schedule, thanks to strong delivery from all products in the program. In particular, strong reliability improvements within production and our procurement excellence product really starting to gain traction. I must emphasize two important points about our improvement program. First, we are investing in long-term sustainable improvement of our operations, including increased production volume, improved energy efficiency, improved variable cost, and lower fixed cost. Second, we're still early in our improvement journey and must continue to work hard for a long time to reach our ultimate goal of $500 million by 2020. While we are proud of the good results that we have achieved so far, we must be prepared for setbacks along the way. For example, within production reliability. We will therefore not adjust our targets based on a few quarters of very strong performance.
At the full year result presentation, we will provide more detailed breakdown of the program results. In the meantime, I'd like to give you a brief update on the rollout of the Yara Improvement Program at three of our locations. In our first quarter results presentation, I told you how the sites at Uusikaupunki and Sluiskil have improved their performance. They continue to deliver strong results, both in terms of savings and increased volumes. For instance, in Uusikaupunki, we have achieved more than 60 daily production records in the last few months, and that's well above even our stretch targets for the plant.
Since then, I have also visited the Belle Plaine plant in Canada, where I saw a strong commitment to changing the way of working, and the results of the commitment can be seen in improved productivity, both at the plant in total and also within individual teams. These teams have shown dramatic improvements, a systematic approach to eliminate idle time and improving planning maintenance in the maintenance department has allowed us to do more work with less working hours and reducing our need for contractors. Also increasing our reliability at the plant. In addition to that, we had a fixed cost reduction at that plant of $1 million in the quarter.
One concrete example of eliminating waste is the maintenance planning at Belle Plaine, where they suggested that they would add a separate section on their performance board, where field technicians can now provide feedback on the quality of the plants after each task. This has led to fewer return trips for missing parts, better task instructions, quality improvements, and time improvements. When the idea comes directly from those involved in the operation in the production, it's much quicker to implement. In a short period of time, the team in Belle Plaine has transformed their site operations, and in my whole career, I have never seen such a positive change in engagement at any plant or site in such a short time.
If Belle Plaine is anything to go by, I'm confident that I will be telling you about similar improvements at other plants in the quarters to come as our rollout continues and the numbers expand. We are inviting investors to our Sluiskil plant on December 12th to see the Yara Productivity System in practice, so please do get in touch with the investor relations if you'd like to join us at this event. Before I hand over to Torgeir, I would like to show a video of the ongoing rollout of the Yara Productivity System, highlighting the very positive employee involvement and results that we are achieving. Rather than short-term cost cutting, we are driving lasting change and improvement and, at the same time, delivering on our mission to responsibly feed the world and protect the planet.
Before I start the video, our program is really gaining traction now, not only within production but within multiple projects across the whole organization. I would like to take this opportunity to thank all our employees for their efforts to get so far in such a short space of time. This is only the start of the journey, but it has been a really good start.
Yara Improvement Program is all about improving the performance of our operations, making sure that we are competitive in long term.
To me, the most important part of this program, and one that I really stand behind, is to get the workforce totally engaged into this process.
We decided not to go for cost-cutting, but to take a sustainable approach towards our improvement by working on the efficiency on our internal work processes.
Our employees, they know their business. They know the work they do on a daily basis, and for them, it is important that they will get the opportunity to say, "Make your job better and more efficient.
Everyone is involved, not only about improving the way we work, but to see that there are better ways to do the work.
I have been collecting over 115 improvement ideas from our personnel. All employees feel that they have a sense of ownership, and their ideas are taken seriously.
Continuous improvement means never be satisfied with the status. Continuous improvement means also the first shot on something doesn't have to be perfect. If it is directing in the right direction, good. Improve. Then you can take another cycle and improve it even further.
In the past, we used to have training, but it was not structured enough. Nowadays, we put more focus on developing what we would call a training package and a structured way of training.
One of the tools of the program is working for us in production is the SOP process. Creating the standard operating procedures gains a lot of educational material for our new operators.
Root cause analysis is a systematic way of working. Very often, we see a problem, we want to fix it right away. That's ingrained in a production environment. If you really do a proper root cause analysis and you find the real issue, it's not only satisfying, but that's also where you benefit dramatically.
One of our NPK operators gave me one very simple idea how to shorten the weekly maintenance stops. We immediately take that in use, and it has already been a success.
I think our mission and vision, feed the world a responsible way. We here in the NPK production play a big role, not only by producing more efficient fertilizers, but also same time making that environmental friendly.
Operating at a higher efficiency by definition means that you are producing more of the same input. In this way, you are basically using less resources for the same output, and that will directly impact on the emissions to the environment.
One of the big improvements in air emissions was this installing of catalyst to nitric acid production, reducing N2O emissions about 90%.
You can see the seeding in the spring, you can see the crop growing, and you know that you're part of that. At the end of the day, you see the big harvest, and you know really, yes, you're part of this big value chain. That's great.
Good morning to all of you from me, too. It's my pleasure then to provide you with some more details on Yara's financial performance in the third quarter. Let me start with the development in our EBITDA. We report an EBITDA in the third quarter of NOK 2,386 million, That is 21% down from same quarter last year. If you then go into details explaining that development in EBITDA from last year, you will see that the single biggest element there is the so-called special items that Svein Tore already commented on. We had a negative special item in total in this quarter of NOK 343 million, of which the largest element is the provision for the closure of the Pardies site in France. We also have some increased provisions for environmental liabilities at an older site in France.
While in the third quarter last year, we had a small positive special item related to contract derivative effect in the gas contract. If you take away those special items, we had a decline in the EBITDA excluding special items of 8%. Let me try to explain that to you. You see here then that the biggest explanation, and fully explaining the decline in the EBITDA, is increased energy cost in the quarter. European energy costs were up 15% from last year, and we also had an increase in our ammonia plant in Pilbara due to the contractual step up in the gas pricing there happening in the fourth quarter of last year. Second negative element compared to last year is the currency development.
The US dollar have weakened towards most currencies globally, and we sell most of our products in US dollar, or the underlying value of the products are in US dollar, while most of our fixed costs are in non-US dollar. A weakening US dollar, about 4% weaker compared with the Norwegian kroner compared to last year, had a negative effect of NOK 214 million in our EBITDA. The third reduction element in our EBITDA, called other, is mainly linked to fixed costs. We are growing, and we are also implementing the Yara Improvement Program. We have about NOK 30 million in higher cost due to the Yara Improvement Program this year than last year. If you look at our total fixed cost, it's up, but it's still up, including the Yara Improvement Program, lower than inflation.
We are beating inflation, but as planned, the Yara Improvement Program has a later traction on fixed cost, as we later are able to take out fixed cost as we improve operations of our plants. If I turn to the positive changes from last year, we have a significant increase in sales volumes. Our fertilizer sales volumes are up 6% from last year, and the biggest explanation there are increased sales of urea in North America. That's partly a phasing effect from the second into the third quarter, but underlying, it's also an effect of a strong production performance in our Belle Plaine plant, as Svein Tore commented upon. In total, it's an effect of strong production performance in Yara. Our finished fertilizer production is up 8% from last year. The second biggest element in growth in fertilizer is related to Brazil.
We sold 4% more fertilizer in Brazil this year than last year. That's also a positive product mix, which also gave a positive effect on this volume, where we sold 15% more premium product, while our volumes of bulk blend were down 13%. Last on our positive volume effects, I would mention our industrial segment, which continued to grow very nicely. Industrial deliveries are up 6% from last year. The biggest explanation there is AdBlue, the solution for NOx abatement from trucks, that continued to grow and were up 15% from last year. Industrial are also able to continue to realize growth in most of their product segments. The last positive element is related to price and margin development.
Some of you may feel that that isn't big in this quarter, because prices have continued to move up in general for fertilizer, or nitrate prices realized in the quarter was 10% higher, NPK prices were 3% higher, and urea prices were 2% higher. If you only took that effect on nitrates and NPK, that could explain a positive margin effect of roughly NOK 350 million. We have some also negative elements here. One is on ammonia sales, where we have a negative margin effect of roughly NOK 100 million, and that is related to positive time lags last year. Urea prices declined from the second into the third quarter last year, we benefited from that in the margins last year.
Another element on margins is related to Brazil and particularly bulk blends there, as volumes dropped there, and also as you had a very strong market last year. This year, our volumes in total in Brazil increased 4%, but last year they increased 11%. The market on commodities in Brazil was weaker this year. Our bulk blend margin was down roughly about NOK 10, which gives a decline of roughly about NOK 150 million related to that. Last, we had a negative margin effect by writing down of damaged goods. We had a couple of incidents with damaged goods, but which we will expect to get coverage from insurance in a later quarter. That had an effect in the quarter of about NOK 100 million. That explains why we don't report a higher positive margin effect this quarter.
If we then move on to see how does this reflect the earnings in the different segments. You can see if you look at production, it is down due to special items. Those I mentioned with Pardies and the environmental liabilities. If you take away that, the underlying earnings in production is on its way up again. That's for the first time for many quarters where we have seen declining commodity prices. This time then, commodity prices helps the production segment with higher nitrate prices, but also higher urea and NPK prices. Clearly the improved production supports the earnings in the production segment with, as I said, 8% up on finished fertilizer, but also importantly 7% up on ammonia. What is then partly taking down the earnings in the production system is clearly higher energy cost, a 15% increase in gas prices.
Industrial continued to deliver very strong earnings, slightly up from last year, and they report in isolation this quarter a CROGI, a cash return on gross investment, above 30%. Crop Nutrition also continued to deliver strong results. They have a CROGI in the quarter of about 13%, but they are down from last year, and that is fully explained by Brazil, by the somewhat weaker margins in Brazil and also a negative currency effect in Brazil as the Brazilian reais is increasing towards the US dollar and revenues are mainly in dollars, while cost in Brazil is mainly in Brazilian reais. The Brazilian effect then is partly offset by better earnings in Crop Nutrition in Europe, mainly due to higher prices there, where as I said, realized nitrate prices up 10%.
If you then move on to some of the other elements in the deviation analysis and start with the energy prices. As I said, Yara's energy prices were up 15%. That was both in Europe, where you saw an increase of the average gas price for Yara up from USD 4.9 per million BTU up to USD 5.6 per million BTU. Outside Europe, you also have the same percentage increase in energy cost, but that was fully explained by Pilbara and the contractual step up last year. The increase of NOK 357 million in gas cost, you can see that NOK 212 was related to Europe. It's included here also North America, but the gas price development in Belle Plaine was flat from last year to this year, and then you have an increase in Pilbara of NOK 161 million.
If you then look forward and look at how could we expect our gas prices to develop the next quarter, and if you use the forward price in Europe currently, we expect that our gas price on average in Europe will increase up to $6.5 next quarter. That will give an effect of roughly NOK 310 million in increased gas cost, partly then compensated by slightly lower or lower gas cost foreseen next quarter in Belle Plaine. For next quarter, we expect an increase in the Pilbara, according to the contract, compared to last year of NOK 100 million. If we then move on to the price and margin development for Yara's most important fertilizer products, I will start with some comments on the price and margin development for nitrates.
As said, we were able to realize 10% higher nitrate prices in Europe this year compared to last year. That was at a starting point, if you compare with the urea prices this year and last year at the beginning of the quarter, they were pretty similar. We felt that we went out with quite ambitious nitrate prices, but we got traction on those, and we were able to realize a nitrate premium in the third quarter on the top of increasing urea prices of $43 per tonne. That is slightly up from the second quarter and significantly higher than last year. Urea prices have increased steeply late third quarter, and we have also increased substantially our nitrate prices in Europe.
If you look at the current nitrate prices in Europe, what we are selling today or taking orders on today and compare that with the current urea price, we have roughly a nitrate premium in Europe on those prices of about $50. We have also said that we have quite a long order book. We have roughly now a three months time difference between when spot prices and when products will be delivered. The reason for that is also, as I said, that we were quite satisfied with the prices we were able to get traction for during the third quarter and choose to sell quite some forward. It was not only us, the total industry in Europe did that as is our interpretation.
That means that when you compare with today's urea prices, which has gone up substantially the last quarter, that will indicate a nitrate premium if you compare with current urea prices and our order book of roughly about $15 per tonne of nitrates. At a much higher urea prices and as such, higher nitrate prices than what we realized in the third quarter. If you then move on to our other most important fertilizer product, NPKs. On top of commodity nitrogen values, NPKs create value for Yara in two steps. First, you could say as phosphate fertilizer, where we dissolve phosphate rock and produce phosphate fertilizer, and that is illustrated on the left side on this slide. What you can see there is that NPK prices have increased slightly from last year, while input factors like phosphate rock has gone down.
That upgrading margin quarter-over-quarter have improved roughly 20%. On the right-hand side, you see the premium that NPK is able to get on top of pure commodity prices, the added value that the farmers are willing to pay for NPKs compared to a bulk blend. That margin, including also a nitrate premium, has been quite stable from last year to this year. We're able to defend and support and develop a substantial margin there. If you look at this table, you will see that NPK price we are able to achieve roughly is 60% higher than a comparable bulk blend on commodities. The last slide I would like to share with you shows the debt development and as such the cash flow in Yara during this quarter.
We started the quarter with a net interest-bearing debt of NOK 16.8 billion, which gives a debt equity ratio of 0.22. We had a cash earnings in the quarter of NOK 2.1 billion. We had a net operating capital release of slightly more than NOK 200 million, and that is related to lower inventories at end of third quarter compared to second quarter. We invested NOK 2.9 billion in the third quarter. NOK 1.1 billion is related to maintenance cost, and NOK 1.8 billion is related to the ongoing growth projects we have in greenfield and brownfield expansions. We have also previously guided on the estimated CapEx totally for Yara in 2017. Last time we guided on a total CapEx this year of NOK 17.8 billion.
You will see in the handout, and the attached slides to the handout, that our current new estimate now is a CapEx in 2017 of NOK 16.5 billion. That reduction is partly due to currency changes, a stronger Norwegian kroner compared to EUR and USD that we actually invest in, but it's also related to some slight delays for some of our projects. The biggest delay, but which is still early in the development, so we plan to catch it up, is Rio Grande expansion in Brazil. The debt is slightly taken down by the currency effects that Svein Tore mentioned over the P&L, but we also have significant currency effects on other debt elements where we have hedge accounting.
The positive other element on taking down net debt there is partly related to currency effect that doesn't go over the P&L, and it's related to non-cash elements in the earnings, like the provisions for closing in Pardies that is not yet cashed out. We end this quarter with a net debt of NOK 16.5 billion, slightly down from the beginning of the quarter, still at a debt equity ratio of 0.22. We continue to have a strong balance sheet for further operation, further growth, and cash return to the owners. With those words, I hand over again to Svein Tore for him to comment on Yara's prospects.
Thank you, Torgeir. In terms of the prospects, I want to provide a further update on the strong urea oversupply in the U.S. and elsewhere, with capacity additions looking set to peak this year, according to the latest update from CRU. As you can see, planned capacity expansions are in excess of trend consumption growth, both for 2017 and for 2018. In addition to the capacity forecast that you can see in the bars up here, we have plotted in CRU's forecast on year-over-year production change. You can see that in the black line on the screen here. This is intended to also capture changes to utilization of existing capacity in the market. There are no major changes to the CRU assessment compared to the previous estimate. However, CRU has increased their production forecast both for 2016 and for 2017.
Looking at other sources, such as the International Fertilizer Association, supply growth outside China was most likely stronger in 2016 than the CRU numbers indicate, due not only to new plants, but also higher utilization in existing plants. As for 2017, the supply increase looks likely to be significantly lower, among other things, due to the closure of the largest Algeria plant due to the conflict between the owner and the major contractor at the plant, to delays on new capacity. This means that there is potential for significant supply increase in 2018. We therefore do not see or expect a fundamental improvement in the global supply-demand balance outside China until after 2018. Chinese urea prices continue to be a key reference point for global nitrogen pricing. Higher production cost in China has resulted in significant curtailments and in reduced exports.
The domestic urea price in China has been relatively stable through 2017, close to the cost of production for the high-cost producers, increasing over the last month as anthracite coal prices have increased further. You can find this data under additional information in today's presentation. Through most of 2017, Chinese producers have discounted export prices compared to the domestic market. You can see this in the blue line and compare that to the black one for the domestic prices. Recently, with low availability in Chinese ports, improved demand for Chinese exports, prices have increased above the domestic price, as you can also see here at the end of the graph, where the blue line crosses the black one. Even more importantly, during the second and most of the third quarter, global prices disconnected from Chinese export prices due to low global demand.
This changed into September, when stronger global demand brought global pricing in line with Chinese export prices. The short-term volatility that we've seen in 2017 can be expected also going forward due to less predictable situation given lower overall demand and supply of Chinese urea export. The outages in Algeria and the good demand from Europe have boosted the North African prices even more than the other references. A further important factor in the global nitrogen balance is import demand from India. So far this season, April through August, sales is up by 3%, while production is down by 2%. In addition, April through August, imports were down by half a million tons compared to the same period last year. Stocks have been declining, India has been active in the market from September, importing more than last year, supporting the market.
As can be seen from the graph on the left, annual sales vary from year to year, partly due to timing of imports. India imported record volumes in 2015 but ended with high stocks, which negatively affected imports and sales in 2016. I want to round up with a summary of the growth investments and the Improvement Program effects. On the left-hand side, we have added together investments we are making both in the Yara Improvement Program and for our committed expansion and growth projects. On the right-hand side, you can see the combined projected earnings improvement resulting from these investments on a 2015 baseline totaling $1.1 billion of EBITDA within 2020, equivalent to NOK 16 net income per share at today's currency rates. For a full overview of our planned investments, including maintenance CapEx, refer to the first three slides under additional information in today's presentation.
With this closing summary, I'd like to hand back to Tor, who will coordinate today's Q&A session.
We are just about ready for the Q&A session. If you have a question, please raise your hand and my new colleague in IR, Nina Kleiv, will endeavor to get you a microphone. Who would like the first question? Should we start with ABG?
Yes, good morning. Bengt Jonassen from ABG Sundal Collier. I have three questions, I think. If I compare the variance table in your quarterly report with respect to the ongoing Improvement Program, how should we read the variance table compared to the $250 million improvements realized so far in the program compared to last year? That's the first question. The second question is on the acquisition in India. I think that was supposed to be closed in August, September this year, and now we obviously are in October. Any development there? I think I'll leave it those two questions. Thank you.
Maybe I start with the improvement program then. You could say the NOK 210 million in total accumulated improvements is based on 2015 prices. You could say, if you take that total split, about half of the improvement is related to production volumes, about a quarter is related to energy prices, and about the last quarter is related to procurement or lower unit cost. The last one is not influenced by, you could say, pricing as such then. That's a relative element. While the two other one is influenced by prices. If you then look at the-- We don't have a split quarter by quarter on this. We plan to give you more granularity on this by the fourth quarter, when we have an extended presentation on that.
To reason a little bit on it then, if you look at year-to-date prices, or if you look at third quarter prices, you will reduce the total improvement with, it will be about 70%-80% of the improvement you have with 2015 prices. You'll take down the production part somewhat, and you'll take down a little bit the energy element. Then, of course, you have to realize that this is annual effects, so for a quarter, you have to divide on four then. What you'll see is in the volume part in the deviation analysis, you have a significant positive element, which is related, of course, to sales volumes, but underlying there, you have quite a lot of the production improvement then, which year-over-year would be in, say, a NOK 50 million on an annual basis then.
I'll give some comments on India. As you correctly point out, our expectation was that we would close this transaction in third quarter. It is taking some more time, partly due to a new regulatory process in India, which has recently come into effect. Both parties are ensuring that we transfer the assets in a satisfactory way, and we're working full speed on our side, and Tata Chemicals is working full speed at their side in order to get this done in the correct way. We are aiming for closing this before the end of the year, but there is some risk that it will take some more time than that to get everything formally in place.
I think we go to Handelsbanken next, followed by Danske.
Thank you. It's Anne-Johan. I have a question related to the fact that you told that you and the rest of the industry in Europe have sold quite a lot of fourth quarter volumes rather early. How is it like today? Do you sell quite a lot as early as now when it comes to first quarter volume? If that is difficult to comment on, what is kind of the normal by the end of October, how much is then sold out of first quarter volumes?
I should start, it's not that much sold out of first quarter, no. You could say both we have a long order book, so we are not in a hurry to sell as such. Also the distributors are quite well covered for their first delivery then. What I can say is that we have started to take order or got orders on the new price level, but neither we nor the customers are in a hurry to take new orders at the current time of the season. Don't know, Dag Tore, if you would say more on that.
I think it's quite normal that, not only for us, but for the whole industry, that you're kind of more keen to try to secure the period until application really starts. When you sell more or less for direct onto the fields, then they are normally less forward selling when you get to that period. Let's say to bridge the fourth quarter, which can sometimes be challenging, depending on the kind of sentiment in the market. There has been quite a lot of forward selling this season as mentioned.
Thank you.
Okay. I think if you just pass the microphone right to the back row where we have Danske Bank waiting to ask a question.
Good morning. Eirik with Danske Bank. I have three questions. One, if you could just explain the reasons, and maybe sensitivity on the adjustment on the expansion projects from 650 to 600, and the kind of sensitivity on if this could be adjusted going forward. Also some on the urea supply side. The CRU slide indicates a higher production change for 2017, as you talked about. Does this kind of skew the picture so that more has come and less to be come going forward? Another thing on the supply side, in addition to the coal price increase in China, do you see any direct closures from environmental reasons, or are they mainly happening on the back of margin squeezes?
I could start with the latter part, whether we see significant impact from environmental point of view. As of yet, I would say that that's very limited so far. Most of the closures are due to financial reasons. I see with the ongoing efforts in China at the moment to reduce emissions, that could provide further shutdowns in the future. As of now, it's mainly driven by financial reasons. I think you continue, Dag Tore, on supply additions, and then you take the capacity.
Yeah. As you said, there is a slight increase in the supply increases for 2016 and 2017 outside China in the CRU slide. I think in 2016, as Svein Tore mentioned, that is called for because the IFA annual complete data that has just been released show an 8 million tons supply increase outside China for 2016, which is even a bit higher than what that CRU estimate is. Part of the reason is the new projects, but it's also a good performance elsewhere. Like Egypt, for instance, added more than 2 million tons in 2016 compared to 2015. A quarter of the growth is from Egypt alone due to better gas supply, because the plants were already controlled before 2016.
That also means that demand growth seems to be a bit better than many predicted, because we don't think that stocks in a way have been kind of fundamentally increased. There's not that much of strategic stocks on fertilizer like on brainstorm or et cetera. A quite good year also from a demand perspective in 2016. When it comes to 2017, we believe that the opposite will happen when we get the final numbers, that CRU are overstating the supply increase outside China. Based on the IFA survey for first half, there is quite limited growth, and Algeria is one example of that. They have cut their exports quite significantly due to the outage of the largest plant. Also elsewhere, they talk about more curtailments and more technical issues and more turnarounds and more different types of things in 2017 than in 2016.
Of course, it remains to be seen how that plays out for next year. As Svein Tore mentioned, there is at least potential for quite a lot of supply increase, but not so much the way it looks today because still delays the new projects and so on. That, of course, hard to say.
Yeah. On your question on the P&L effect of the expansions, and you're referring to, you said, the last slide of Svein Tore's presentation, where it shows that you expect that the expansion project will have an EBITDA effect of $600 million by 2020 or six kroner per share. You're totally right that previously we have guided on $650 million and seven kroner per share. The reason for that is, if you look at the timing of the project, I said that we had some delays on the Rio Grande project, but we expect to catch up that. We also see somewhat later finalization of the Salitre project. We expect, as we did a quarter ago, that the mining part of that project will be ready in first quarter next year.
We previously guided that we thought that the chemical part of it, where we take the rock and upgrade it to phosphate fertilizer, would be ended by end of 2018. Now we see mid-2019. There will be a later step-up of production in that project, which influence the earnings expected in 2020. That's the major element of that.
If you included 2021, we would probably be at the same number?
Yeah, that's right. It's a slight delay in step up.
We have the next question from DNB. You bring the mic forward again or via the side to optimize logistics there.
Thank you. Eivind Løving, DNB. Two questions on the markets. One on the nitrate stocks. You mentioned a couple of times that the producers are comfortable. What would that indicate for your sales in Q4? Also on India, the import demand, where are the current Indian stocks, and what does that indicate for imports for the rest of this season and also going later into 2018? Thank you.
You'd like to start with India?
It's hard, as everywhere else, to judge the exact inventory positions because how much is on farmer, distributor, producer level. If you just look at the development since the start of the new year in India, which starts April 1st, there is roughly a 1 million ton stock reduction implied by if you just take sales minus production and adjust for imports, roughly 1 million tons. Of course, we don't know exactly what consumption have been and so on. What we see is that they, through two tenders in September, they bought basically 1 million tons already, supported the market. There is a tender ongoing now, they don't seem to be able to buy more than three, 400,000 tons in that tender. At least so far, that's the indication. Just because there isn't much availability at that price.
The traders, they find some Iranian urea, they find some Chinese, nobody else seem to be keen to sell at USD 290 CFR India. I guess there's already speculation in the press that another tender is just around the corner or not far away. India is certainly a very supportive factor in the market right now, and I think reducing the risk of, let's say, temporary setbacks. Of course, what happens in 2018 is largely then depending on what the consumption developments in India, which are, of course, more uncertain also with, now they are going to try to reduce the bag size from 50 kilos to 45 kilos. There are also the direct money transfer to farmers so that farmers trigger the subsidy payments rather than the sales from the producers.
These effects are a bit unclear, at least they are not doing anything with the retail price, it seems like. Urea will still be very cheap, below USD 100 for farmers. There are no signs of doing anything with that. I would say that the consumption developments is then the key to what happens further on after this, let's say, fourth and first quarter, where it certainly looks stronger than a year ago.
Yeah. On nitrate deliveries, as you allude to, of course, when we have a long order book, we have much more certainty on deliveries. There are still facing uncertainties there when we choose to deliver or when the customer choose to take order and the flexibility of that. That can vary a little bit if those end before or after the quarter. We are in a very comfortable situation, of course, when it comes to the order book. There are also, of course, uncertainties on deliveries outside Europe, where we have a growing market, particularly in Brazil, and where not all is booked for the fourth quarter. The fourth quarter is also, in a way, the end of the season in Brazil. There are some uncertainties there.
Another element is also maybe just pointing to the inventory situation in Europe, which is quite normal. We have one slide in the attachment there, slide 35, or sorry, 36, where you'll see that the estimate of the current inventory situation in the European industry is on average of what it has been over the last five years, but actually lower than it was the previous couple of years.
Are there more questions? There is another opportunity, for those who are interested, at 2:00 P.M. Oslo time, we have a conference call. Until then, thank you for your attention and attendance. There are more prints outside for those of you who didn't get one on the way in.