Good morning everyone, and welcome to Hydro's Capital Markets Day 2016. My name is Stian Aarsæther, and I am together with my colleague Olena Lepikhina, responsible for investor relations in Hydro. We are very happy to welcome you here today to the London Stock Exchange, and we believe we have a very interesting agenda for you today. Going through the main developments and achievements, since our last Capital Markets Day, and also talking about our strategies and priorities going forward. This is the second year in a row we are hosting our Capital Markets Day here in London, and as last year, we are very happy to see so many of you have taken the trip from different places around Europe to join us here today. I also know quite a few are following us on the webcast, so welcome to you as well.
Before we do anything else, we would like to start with safety. There are no scheduled fire alarm tests, today, so if the fire alarm goes off, please listen to the announcement, and follow the instructions. The fire assembly point is over by St. Paul's Cathedral, and the event management team will be on hand to assist. The nearest fire exits are behind us, out the entrance doors and turn left or right and follow the green signs. In the event of a medical emergency, please contact the event management team. Also before we start, I would like to direct your attention to the cautionary note related to forward-looking statements, which can be seen on the screen and also in your presentations. Let's move over to the agenda, which looks as follows.
We'll start off with the CEO, then the CFO, and then a Q&A. After the short break, we will have a market update followed by Rolled Products and Primary Metal, and then a new Q&A. Then we have the final break, followed by presentations from Bauxite, Alumina, and Energy. This is followed by a new Q&A, before a final summary and Q&A with our CEO. There will be lunch from 1;00PM to 2:00PM,, and I hope as many of you as possible can join us for that as well. With that, I will leave the stage to our President and CEO, Svein Richard Brandtzæg, but first, let's take a look at this.
Consumers do more than consume. Consumers make choices. Consumers care about their family, their future, the planet. To live up to these expectations, companies too must care, and be able to show how their products are made in ways that harvest the fruits of the world without consuming the world itself. In Hydro, we will meet future expectations by keeping control over our products in every step of the way. Our competence stretches beyond and between processes, and with the entire value chain in-house, we are able to safeguard that every single atom leaving our production facilities, contains the values and the responsibility that is the Hydro way. That's why we say it is not aluminum, it is Hydro-, aluminum. The energy we put into our metal is not consumed or lost, and by being infinitely recyclable, the aluminum can be used over and over again.
By mastering all the steps of the value chain, we can use our deep understanding of natural science to manipulate aluminum atoms so,, that our metal achieves the specific qualities we aim for and consumers care for. Using Hydro's large share of renewable energy, together with our unique smelting technology, we are taming the forces of nature, getting more aluminum out of the energy we put in, and reducing emissions. With our top-quality alumina, we are able to produce quality metal with low environmental impacts. To make aluminum, we need the raw material bauxite, a raw material found in the ground around the equator. When we extract bauxite from the ground, we are at the very first step of our value chain. To ensure that we maintain the biodiversity in the area of the mine, we replant the forest in line with the natural habitat of that area.
Again, to ensure that when something is gained, as little as possible is lost. By being a 360degree aluminum company, we are in a unique position to control and be responsible for the land, water, and forests, for our employees and the communities we engage with, for the energy use and emissions in our processes, and for bringing end products back into the loop to be used over and over again. Hydro- infinite aluminum.
Yes, to all of you here at London Stock Exchange, and also to those of you following us on the webcast, welcome to the Hydro Capital Markets Day 2016. We have had quite an exciting year, since we met here in London last time, and all indicators shows that there will be also an exciting year ahead of us. We are now the only pure play integrated aluminum company in the Western world, and I'm convinced that this is the best way for us to serve our customers and also to generate shareholder value. Keeping in mind the integrated business model of Hydro, there are four segments I will cover in my presentation. First of all, I will give you an update since the Capital Markets Day last year, the main highlights since we met here last time.
Secondly, I will discuss the world around us and the market situation in our industry. Then thirdly, the response in Hydro and how we are becoming better, bigger and greener. Finally, the main priorities and the direction of the company. In line with our philosophy, that responsibility is on top of our agenda, I would like to start with the health, environment and safety and social responsibility. Let's start with safety, where we are now approaching the end of 2016, which is coming out as the safest year in Hydro's history, if we continue at the same pace on safety performance as we have done during the first months this year. We still have December left, but the safety performance so far is the best in our history.
That is, of course, an important achievement in itself because it's important for us to take good care of our employees. We also are very well aware of the close relationship between operational performance, and safety performance. Also, with regard to fatalities, this is something we really want to avoid, and there is quite some time now since there's been a fatality in Hydro. We have the most ambitious climate strategy in the company to become climate neutral from a life cycle perspective in 2020, and we are on track. We are also on track with our environmental target, to have one-to-one reforestation in Paragominas, and also that target is on track. With regard to corporate social responsibility, we are listed on the Dow Jones Sustainability Index, as have been there since 1999.
We are a member of the UN Sustainability Stock Index, the Global Compact. We are the proud founder of the Aluminium Stewardship Initiative, and we are named as business frontier on the UN Sustainable Development Goals, and we are also active member of the International Council on Mining and Metals. In summary, it's a strong indication that we are on the right track, 2016 has been a year with strong progress in the company, and let me go through some of the main highlights, which is shown in this slide. First of all, I'm very happy that we have achieved record production in our operations in Brazil with record high bauxite production and record high alumina production. That implies also the cost position, which is also now record low.
If you move over to Primary Metal, also, there the improvements continues, and we are now in the end of the year going to deliver the $180 program that we had talked about previously. We also are very happy that we are now constructing the technology pilot at Karmøy, which will be 70% finished in the end of the year. In Rolled Products, we are ramping up the used beverage can recycling line in Rheinbach and also ramping up the automotive line number three in Grevenbroich. We have also finalized the debottlenecking of the world's largest hot mill, Alunorf, in Germany. In Energy, we continue to secure power contracts long term and that is an ongoing effort in Energy.
We are very happy that the parliament decided on the change of industrial ownership of power after reversion, which is a small change in the reversion legislation in Norway. That will secure the value of all the hydropower assets in Norway even after reversion. In Sapa, it has been a record year, record performance, and with improvements in several areas. Hydro has also signed a long-term agreement on supply of extrusion ingots to the joint venture, Sapa. I show you this last year where we have listed the long-term goals for Hydro in several areas, again, to become better, bigger and greener. There are strong improvements in all areas. As you see, most of the lights are green.
There is one red light, and that is regarding the equity bauxite target, where we did not agree with Vale this year to acquire the 40% share in the MRN mine. We have the evergreen agreement with them. We are also somewhat behind on ramping up the used beverage can recycling line in Germany, so that is yellow. We are working with that. That means we are also behind on the short-term target on recycling volumes. Long-term, we think we should be able to catch up with that. All in all, strong progress on several areas, and it also shows that the improvements we have set up as targets are also moving forward as we had planned.
Doing well in Hydro,, is very important from an absolute perspective, but it's also important for us to have a relatively strong position in the industry. This is now curves taken from CRU showing our position on the cost curve, on smelting cost curve, and the B&A on the alumina cost curve. As you see, we have fairly strong positions. This is again position influenced by currencies that we cannot influence, but it's also very strongly determined on the improvement programs that we have delivered on during the last year. For us, it's important to continue with our improvements to keep the competitive position. There are also other areas we are comparing with our peers, and for example, on CO2 emissions per kilo aluminum, we are the best in the whole industry.
As you see, we are on the lowest part on the right scale here. This is a strong position for us going forward. Also, with regard to safety, this is the numbers from International Council on Mining and Metals from 2015. Also there we are among the best in the peer group. We, of course, will never finish our work on safety. That is a never-ending work for us to work with safer operations. If you then move over to financials, and look at balance sheets and compare with our peers, we have the strongest balance sheet and lowest gearing among our peers in the industry.
I believe this is very important for us to have a strong financial position in a cyclical and quite volatile industry, as we have experienced during the last years. If you also look at the underlying payout ratio, and the dividend, we are also the best in class among the peers in our industry. If you then take a look at the total shareholder return, now take a period from 2008 to 2011, which includes the financial crisis, this is highly negative. We are best in class, but on a negative scale. If you then look at the period from 2012 to 2016, 2015, it looks better.
We are also here best in class, but also here, I must say I'm not happy with the level of total shareholder return, which is very much related to the situation for the whole industry, where the profitability has not been at a sustainable level for quite some time. If you look at profitability in our industry, we see that it's now improving, and Hydro is ahead of the average profitability in the industry. Still, we have a way to go before we are coming up to sustainable profitability levels, which is, again, a strong motivation for us to continue our efforts in improving our businesses along the value chain.
Now that we have looked at where we are today, both in terms of our goals and ambitions compared with our peers, let us then move to what is the priorities for us going forward. Our three main elements here that I would like you to take as key messages from my presentation, which is firstly, related to continue to manage the cyclicality in our industry through the financial strength and flexibility, meaning a strong balance sheet going forward. The second part is, related to the improvements that we'll continue and have improvement programs very high on the agenda, in Hydro going forward. This is very much about cost-cutting, productivity improvements, but also about high grading our product portfolio.
Finally, we will continue as an integrated company, differentiate ourselves from the rest of the industry, and again, living up to these ambitions are at the core of my obligations to Hydro, to the market, and to the shareholders. Let us then move over to the world around us, and look at the challenges and opportunities in the aluminum industry going forward. You're very well aware of the macroeconomic situation, and also the geopolitical risks globally. Let me give some comments to this. Aluminum is still a function with regard to demand, a function of the GDP development globally. We see in emerging economies that the speed is changing. China is now slowing down. India keeping a high pace, and Brazil, where we are operating, is now showing some improvements.
With regard to trade and framework, there are some concerns after Brexit and the US election. That is something we are following carefully. With regard to climate change and the challenge about the CO2 emissions, this is now becoming a wider agenda. It's coming higher up on a wider community. It's not only regulators and NGOs. That is something we will discuss further also in my presentation. Finally, with regard to the digital revolution with new technologies, I would say that for Hydro, this is not a revolution. It is more an evolution, as we have several examples where we already are in Industry 4.0.
I will come back to that also later. China is so important for us that I would like to comment on the development in China, where we have now seen the demand is going up to a level corresponding to about 50% of the global consumption of aluminum. Similar, the production capacity is now at least 50% of the global capacity. This is very much based on the general GDP development, the demand growth, building infrastructure, cities, buildings, and also, of course, the fact that there has been easy access to capital. Environmental regulations are not the same as we are used to in other parts of the world. It has been a quite easy access to energy, especially then coal energy.
Also, China has developed their own raw material basis, gradually the whole time, and this is now changing. There are several factors here that are changing. First of all, the growth is going down, which is then also what we can see with regard to the appetite to build new capacity, that is reduced. We see financing is getting more tight. The governments are more concerned about the environmental situation in the cities. Also with regard to energy, we see coal prices are increasing, energy becomes more expensive. Also on the raw material side, China is now on the way to deplete their quality bauxite reserves. There are still bauxite in China, but that has very high reactive silica content and low alumina content.
That means it's very more and more expensive to utilize these reserves, so China becomes more dependent on imports from other countries on raw material. Going forward, we expect that China will continue to maintain its policy of being balanced on primary metal, meaning they will produce as much as the demand in China will be. That is what we expect. They will also continue to utilize the opportunities for exports of semi-fabricated products, semis, when the opportunities are in place. Again, China will still play a very important role for us, and we continue to follow China carefully going forward. If China is a facet of the global mega trends, then I would say that corporate responsibility is also that.
I've already talked about corporate social responsibility, which is not only on topic for regulators and NGOs. We see our customers are now also concerned about the value chain, how are we operating along the whole value chain. The car producers that I met recently, two days ago was the last one, that are talking about how can they supply sustainable material for their cars. It's not only the emission of cars and the fuel consumption that is important for them going forward, but it's also the carbon footprint of the car before the car enters the road. There we are in a very strong position. With regard to international finance, I meet some of you in our quarterly roadshows.
I get more and more questions about social responsibility, environmental performance, safety performance, and the extra financial dimension, that is now being more and more coming closer to the financial dimension. In Hydro, we are welcoming this development. We think this is very good for aluminum as a metal, and it's especially good for Hydro due to the fact that we have a strong track record in this respect. Let us move over to the expected demand development going forward. Now, we are taking the demand expectation long term a notch down compared to what we have talked about previously. We are expecting long-term demand of primary to be about two to three percent, and now we are talking about demand towards 2025. Short term, we expect a three to five percent growth globally.
That is the expected growth for next year. That will be on higher level. The demand of primary of two to three percent is based on the semi demand of three to four percent long term and also increased recycling growth four to five percent in recycling. These are the long-term growth figures, but there are quite some interesting sectors and market segments that are growing at different speeds, and let us take a closer look at that. Here we have an overview of some of the main markets for us, transport, automotive, packaging, building construction, and electrical. As you see, some of these segments have much higher growth than the average long-term growth.
Of course, it's depending on the market segment itself, but we're also talking about now substitution at a higher pace than what we have seen before. In automotive, it is very much about exchanging steel with aluminum. In automotive, there are also on the electrical side changes with regard to cabling, of course, where copper cabling is moved over to aluminum cabling, as we also see in other electrical applications. Packaging is also again, it's a growing business in itself, but also there we are benefiting from the fact that steel packaging is moved and steel cans are moving over to aluminum. Building and construction has the lowest growth rate as we expect going forward. Here we are seeing the advantages of aluminum with regard to energy-efficient building solutions.
We are well-positioned in these areas, and aluminum is the metal of choice for many of these segments due to the attractive properties. The formability of the metal is quite unique, which means that the production processes is very productive and cost efficient. That is again, due to the atomic structure of aluminum, which is a face-centered cubic structure, which is similar to gold and silver, by the way, so we are in good company. Also the anti-corrosion properties of aluminum is very unique. It protects itself with a nano-thick layer of aluminum oxide, in fact, which is transparent, so you can see through it. Aluminum has also good heat conductivity, about 60% of the heat conductivity compared to copper, and also good electrical conductivity.
In fact, also electrical conductivity is about 60% compared to copper. All these properties is a big advantage for our metal, of course, and that means also that the aluminum is growing into more and more segments where there are other materials that has been used previously. We should also, by the way, remember that the aluminum is endlessly recyclable metal, so that's also another benefit. To go to automotive, we are expecting 15% growth in body in white materials going forward. This, the driving force behind this is of course that the OEMs wants to reduce emissions and fuel consumption.
We see this is a significant factor going forward, and that is also the background for our investments that we have done now in Grevenbroich with Automotive Line Three. We see OEMs are very eager to get more metal from us. I have had meetings with some of the biggest OEMs in Europe. We know we are one of the biggest suppliers to the most advanced customers there. It is not only the fact that they want the volumes, but also they are very eager to get hydro-based aluminum produced from renewable energy, to get a lower carbon footprint of the car. As I mentioned, that is becoming also a concern for the automotive producers.
If you now take a look at the demand development, and compare aluminum with the other base metal. Aluminum is the fastest growing metal, and has been so for quite a while. We have included the financial crisis here, and you see aluminum is leading in this growth competition. Also, if you take a look at the growth rate as a function of maturity and economic development, the area of these circles are the accumulated semi- demand per capita from 1950 to 2016.
Just to show you that even China is a big market and have developed a big aluminum industry, it is a long way for China, and as you see also for India, before they are achieving a maturity situation as we have in Europe or in the U.S. market. Meaning that this is still a long way to go for China that is now developing from a economy that has been very dependent on export and in investments, and now moving over to consumption-driven economy. There's still a lot of room for aluminum in a consumption-driven economy, and that's why we are optimistic also when we go to further growth in aluminum in these countries going forward.
If you then take a look at the growth and the supply situation, again, very high growth rates of supply of capacity in China. The gray areas here are related to the period from 2000 to 2016, and then the expectation for growth in demand and production capacity going forward in blue. You see that there are quite some substantial changes here. China that has been on the highest growth curve is now coming down to levels that is also similar to what we are seeing in the Middle East, which is another area where there's been built up significant production capacity of aluminum.
India is continuing on quite high pace, and outside China and India in the Western world, there will be very moderate, very little additions of capacity going forward. We should remember that in the Western world, there are a significant deficit of aluminum already today. If you then take a look at the supply demand balance, and this is a figure I showed you last year that was an update from December 2014, where we showed you a quite balanced market up to 2016 and then an under supply situation.
I gave you an update in the Capital Markets Day 2015, which looked like this, where we expected an oversupply in 2015, balance in 2016 and then undersupply and deficit in 2017 and onwards. Now we have done an update of this based on the information we can get from the outside. We are now expecting again a deficit this year. As you see, the surplus last year was lower than what we expected. We expect the deficit this year, that is what we see now, and Cathrine will come back to some of the details here.
In 2017, we are expecting a small surplus due to the fact that we haven't seen all the effects yet of the restarts of curtail capacity in China, and also they're coming up some new capacity in China. We are expecting a slight surplus in the global industry due to that. From 2018 and then onwards, we are expecting a deficit. We should also then remember that there are global inventories that is registered, and some of them are not registered.
We have tried also to collect all the information that we can get regarding the global inventories, and we see that when we add up the numbers, we see that it's adding up to about 12 million tons globally, which is somewhat lower than what we have communicated previously. Cathrine will come back also with the details on the global inventory level in her presentation. This is again numbers based on CRU at the different points of time. Let me then move over to the response from Hydro, and the direction of Hydro going forward, and then start with the integrated value chain. I'm convinced that this is the best way for us to generate shareholder return.
It is a complex model, but the secret behind this is really to develop different business models, business systems along the value chain. It is very clear that Kjetil is operating his Rolled Products business here very differently from Hilde's Primary Metal, and also from Bauxite & Alumina. By having different production system, business system, we are dedicating the governance system to the market, the competitive environment and the situation in the business area. We are continuing to pursue a full integrated value chain strategy, because there are several advantages with regard to operations and technology. With regard to improvements, for example, we take the key learning points from one business area to another.
We are transferring experiences on the production system from one business areas to another, and we have a unique technological competence and capabilities with having detailed knowledge of the whole value chain of aluminum. With regard to customers and markets, more and more customers are becoming concerned about value chain stewardship, and with full control of the value chain. We also have deep knowledge about that. With regard to innovation, we have several good examples where innovation that happens in the interface between Hydro and the customers are very much dependent on knowledge about the whole value chain. It gives us also a unique market intelligence, the market understanding, due to the fact that we are heavily exposed along the whole value chain of aluminum.
With regard to growth and exposure, of course, with this long value chain, we have a wider range of opportunities for us to grow the company than it would be if we had only have been focusing on one part of the value chain. If you look at the extra financial dimension, it is also an advantage for us to have full control of the value chain, due to the fact that this is now going to play a more important role going forward. With regard to value chain control, with regard to the questions about climate and sustainability, and also compliance and responsibility going forward.
If you then take a look at the investments, and I'm sure that this is very high on your agenda, regarding Hydro's high CapEx level for the moment, and it I can assure you that this is also very high on my agenda. There are good reasons for doing it. We are realizing some projects, where we can increase our value creation potential, and we are growing in specific areas and also have some sustaining CapEx level now that is adding up to the total CapEx level that is higher than what you have seen previously. The Karmøy Technology Pilot is one example, where we are going to test out world-leading technology for aluminum production.
In itself, we are going to produce aluminum with the world's lowest energy consumptio and also lowest emissions. We are also going to test out several technology elements that we can utilize in the existing smelters in order to continue the cost reductions, after we have finished the NOK 2.9 billion program. In Automotive Line Three, we are going to take a number two position in Europe in body-in-white sheet for the automotive market. It is a big step for us, and we see customers are really waiting for the capacity to come on stream, and we are now ramping up this production line, which had the price tag of EUR 130 million. UBC Restarting Line is also on the way to be ramped up.
That had a price tag of EUR 45 million and is somewhat behind on ramping up. With full production there, we will take 20% of all used beverage cans in Europe and recycle that and take it back to the value chain. In Brazil, we have some heavy investments in tailings dam and the bauxite residue deposit. Here we are utilizing the best available technology, all in all, BRL 1.6 billion in Brazil. Again, this can be regarded as a sort of sustaining CapEx because that is something we do every 15, 20 years. When this is finished, it will take some years before we are going to do it again.
All projects are moving according to plan, and time and budget, but we are somewhat behind on the used beverage can line in Rheinbach in Germany. With regard to market positions, we are strong in several areas. In the Rolled Products, for example, we are number one globally in litho and foil. We are number three in can, we are number one in special products, and we are now taking a number two position in body in white for the automotive industry in Europe. In Sapa, we have a number on position in this joint venture in the US, number one position in Europe, with 27% and 24% market share respectively. Sapa is also a global leader in precision tubing and multi-port extrusion for the heat exchanger market.
In Metal Products, we are number one in Europe as the biggest aluminum producer for the local markets in Europe. We have strong position both in U.S. market and also in Asia. With our hydropower production in Norway, we have a strong presence in the Nordic market. We are the second-largest hydropower producer in Norway, and this competence is also quite unique for aluminum company, where we are sourcing energy complex globally for our full value chain. In Brazil, we have excess bauxite and excess alumina, and we are number three globally on the bauxite and alumina supply to the global market. Strong positions globally that we are benefiting from going forward. With regard to advanced products, that is something we continuously develop.
We have, in fact, for several years had a full and increasing focus on research and development to develop specialty alloys for special customers. For example, crash resistant alloys, which is used heavily in automotive industry to make safer cars, is one specialty that we have developed. Aluminum has the special property to absorb energy better than any other metals. If you drive in an aluminum car with crash boxes in aluminum, you have a much better chance to survive if you are unlucky to have a big crash.
We are developing specialty products for heat transfer, like multi-port extrusions, as I mentioned. They are producing this in Sapa, and the alloys, they are developed in Rjukan, which are quite demanding alloys for special purposes, used in air conditioning system and radiators for the automotive industry, and now also moving gradually more and more into the non-automotive heat exchanger businesses. In Rolled Products, we have special alloys, body in white, for the OEMs, and we are the only one that can deliver alloys to our customers, where they can produce and stamp a whole side panel in one piece. That saves cost, but it's very much due to the unique formability that I will come back to later.
We are developing specialty products for special purposes like the Hy-Can product, which is a product that is used for heating of floors, for example. In Extruded Products, we are supporting Sapa also to develop their products, crash-resistant applications for and also advanced profiles for energy-efficient building solutions. This is an area for us that we continuously are developing, and we are doing a lot of work together with our customers, and innovation here is very much about the good cooperation that we have with our good customers in Europe and around the world. Recycling is becoming more and more important for us. Roland Schöppich, will come back to that.
We have a target of 250,000 tons recycled material in 2020. This is a combined strategy from Rolled Products, where we are increasing recycling through the used beverage can recycling line in Germany, and also where Hilde, in Primary Metal is transferring gradually the remelters to become recyclers. Meaning we are not only remelting process scrap, which we are doing today, but also taking in more and more post-consumer scrap into the value chain. This is a good combination where we are also developing new technologies for sorting of metal, that we have recently done. We can take cheaper and cheaper scrap and sort it out. That means we get access to lower cost aluminum and produce higher value-added products. For example, recycling-friendly alloys that is now representing 50% of the volumes from our remelters. This is an interesting business for many aspects.
We get hold of more and more metal units. We are creating higher and higher value out of this and develop new products that the customers are asking for. We have the most ambitious climate strategy in our industry to become carbon neutral from a life cycle perspective in 2020 is quite ambitious. Here we are taking a life cycle perspective, which includes the emissions we have upstream, the benefits of aluminum in the use phase, and also taking into account the benefits of recycling that we are increasing.
In the recycling of aluminum, we need only five percent of the original energy content of aluminum, so that is quite energy efficient, and that is also the reason why 75% of all aluminum that have ever been produced globally is still in use. It's a recycling-friendly material. We're also doing investments according to our strategy. The technology pilot in Karmøy that is really based on the fact that we want to reduce the energy consumption of aluminum production and reduce the emissions from aluminum production. Automotive Line 3 is again to increase the volumes to help our customers to reduce their emissions, and also the recycling line is a good example for us to increase the capability of recycling scrap, secondary scrap from the market.
It is again a part of the climate strategy for us when we are doing these investments going forward. We have had a high activity in research and development, as I mentioned. We have continued to have very strong development and also, I would say, good results from our research and development activities. That is not something we only do in the downstream area, we also do it in other parts of the value chain. We have taken technology leadership now with regard to handling of bauxite residue with the press filter technology that we have developed in Brazil.
I was there when we tested this, and it was impressive to see the compaction, and the concentration of solids, which means that we need less area to store these solids, and also it reduces the cost of storage area. There are clear environmental benefits of this technology. In automotive, we have developed technology for surface treatment of body in white material in connection with the Automotive Line Three, where we are establishing micro cavities in the surface of aluminum, which gives access to lubricants before the OEMs are forming the final products like body in white products for body in cars.
This improves the formability significantly in combination with the changes that we have done on the metallurgical part. These are quite unique materials that gives also new applications or new possibilities for applying aluminum. That is again, a strategy for us to be in the forefront especially in industries where we have very demanding customers, but also where we see higher value creation potential. In the Karmøy Technology Pilot, we have several examples where we are moving towards Industry 4.0. We have already developed predictive computer control, which is very similar technology as meteorologists are using for the weather forecast for next week. We are doing that to control our electrolysis process to control the inter-pole distance in a better way to reduce energy consumption and reduce emissions.
We are using artificial intelligence. We are using new sensors to collect a huge amount of data. We also are developing the capabilities for analyzing big data in our system, again, to reduce the cost of production, to reduce the energy consumption, and reduce the emissions. This is good for Hydro and good for the environment. If you then take a look at the performance improvements that we have worked with in Hydro for quite some time, and the continuous efforts to deliver on improvements, we launched NOK 2.9 billion program to you last year. That was on top of the NOK 4.5 billion improvements that we have delivered up to the end of 2015.
We are now moving according to plan, both with regard to the long-term target of NOK 2.9 billion, and also the yearly target of NOK 1.1 billion this year, with some deviations between the business areas where we see Bauxite & Alumina is ahead of plan. Rolled Products is somewhat behind due to the delay in used beverage can line and also some product mix effects. Primary is also somewhat delayed from accumulated level, but not on speed, but that is due to the power failure in Årdal earlier this year. In total, we are on plan for 2016, and also for 2019. We are talking about assets and improvements and the markets, but we should not forget that Hydro is about people.
We are people that are dedicated to improve the processes, and are dedicated to lift Hydro to higher levels every day. We are building the competence in the organization through the way we are cooperating in our organization. Also, the way we have built up the production systems. I'm sure that Hilde will talk about that in her presentation, where people are cooperating and finding new ways of improving the processes every day. Making sure that what is done today, is better than what was done yesterday. It is a competitive advantage for you, though, and after many years of cost cutting, restructuring, demanding productivity improvements, we could expect that the employee engagement should be reduced.
What we have seen, and we are measuring employee engagement every second year, and, during last measurement that we did this year, in 2016, it was the highest employee engagement that we have ever measured. From a global perspective, we are now among the 10% best in the world with regard to employee engagement. I think all of you are very well aware of the close relationship between the performance index and also the employee engagement index. There's a close correlation here, and, I'm quite convinced that the engaged workforce in Hydro will continue to make Hydro better also in the future. Sapa is also doing a good job.
Sapa has a record year so far in 2016, taking important market positions, high grade in the product portfolio and also continue to move into areas where added value creation can happen. We are very happy for the decision that we made in 2013 to move the Hydro Extrusion businesses, together with the old Sapa Extrusion businesses to create the number one global extrusion company. If you then come to the strategic direction and the priorities, I would like to share with you some of the main points here. We have talked about our efforts to improve our cash flow and emphasize the importance of having a strong financial position.
Many of you may be wondering what will Hydro do with the money. I can assure you that the net cash position of Hydro is not burning a hole in our pockets. We will be disciplined and patient until we find the right target, and the right opportunities that can create long-term value creation for our shareholders. The priorities for cash also depend on where we are on the cycle. Of course, for us, it is high priority to maintain a strong balance sheet, to be able to give you a reliable and predictable dividend. We also are quite committed to keep the quality of our assets at a certain level, to maintain the productivity, and the cost level and be able to improve our relative position in our industry.
Sustaining CapEx is still something that we will look for. There will also be possibilities for debottlenecking small growth investments. If you look more longer term, then of course there could be opportunities, M&A opportunities. There could be organic growth opportunities for us. You probably know there has been talks about CAP II, the CAP project in Brazil. This is not the time for us to build more capacity in alumina and not the right time for us to build more capacity in aluminum. In a situation where there will be excess cash and we don't see opportunities in M&A or organic growth, then we will look into the toolbox and see if there should be share buybacks or special dividends.
That is something that, of course, is going to be decided by or recommended by the board and decided by the general meeting, and that is normally happening in May every year. Again, over longer term, we will continue to keep a strong balance sheet. We will continue to reinvest in our assets to keep them on a decent, good level. In absence of profitable growth opportunities, we will look into the toolbox, what we can do for our shareholders going forward. I showed you a roadmap to profitability last year, and let me give you an update now, which is very much about the profitability of average capital employed. Here, I can show you the development with and without improvements.
If you haven't done the improvements we have done during the last years, we would have a profitability with the current price level $1,700 per ton aluminum $850 and then $300 on alumina PAX level. We would be able to deliver a bit above three percent profitability on capital employed. The delivered improvement efforts that we are going to finalize this year adds up to NOK 5.6 billion. That will give us a profitability of about 8%. If you then add up the next level, which is not the end of the improvements, by the way, but the next level of improvements will take us above 10% profitability at the current price level. This is just to give you an indication of the potentials going forward.
Again, this could be a story of a company in a completely different situation if we haven't delivered on the improvement programs. This is very encouraging for us to continue the improvements, because we see that what we are doing is working significantly when we look at the profitability. Let us then look at the messages I gave you in the beginning of my presentation. I trust that we have now covered the main topics sufficiently, to specify what I would like you to take away from each point. Under managing cyclicality, it's very much about maintaining the financial strength, and continue to lift cash generation and also to maximize shareholder return and value creation for the shareholders. This is the first part of my message.
The second part is very much about the improvements. We will definitely continue our improvements, and we have long-term strategies for that. I'm very happy that we are also moving according to the long-term and also the short-term plan for the improvements. This is very much about cost cutting, but it's also about high grading on the product portfolio. Finally, we will continue as an integrated aluminum company and differentiating through the value chain. All these efforts are very much about the overarching goal for us to improve performance and driving sustainable shareholder value going forward. These are key messages from the Capital Markets Day 2016, and I hope you can bring that with you home.
That again concludes my presentation, which shows that we are on the way to become better, bigger and greener company. Thank you very much for your attention. With that, I give the word to my CFO, Eivind Kallevik. Please, Eivind.
Thank you.
Thank you, Svein Richard Brandtzæg. In my financial update today, I would like to give some comments on the financial developments that we've had since our last Capital Markets Day. I will also talk about some of the key developments in our key financial framework, and discuss how we think about capital management and capital allocation principles in Hydro. We will also look at sensitivities as well as earnings and returns in several different pricing scenarios. Then finally, I will end with our financial targets, as well as aspirations. Let me start with an overview of Hydro's financial performance for the past six years, including the last four quarters. The underlying EBIT and ROACE has been improving since 2012 and reached a current peak in 2015.
The improvement trend has been supported by recovering prices as well as currency tailwinds, but it also reflects Hydro's relentless work to do improvements over time. On the back of the lower realized prices in the last four quarters, earnings and capital returns have come down from 2015 to an underlying EBIT of NOK 6.2 billion and a ROACE level of 5.4% for the last four quarters. The best way we believe to protect ourselves against the earnings volatility in our cyclical industry, is clearly to maintain a solid balance sheet. We have had a net cash position throughout the years, which we've reached just above NOK 5 billion for the last two quarters. The gearing level, here expressed as adjusted net debt to equity, has also been very low, below 30% for the entire period. Maximizing cash is certainly also a priority for us.
The free cash flow has been positive, with the exception of 2011 when we acquired Vale's Aluminum assets. In the latest few years, we've seen a somewhat elevated CapEx profile, which temporarily puts a pressure on the free cash flow. However, the free cash flow has remained positive both before, and even after the dividend payments that we do every year. Finally, everything that we do is intended to maximize long-term value creation for our shareholders. While total shareholder returns have been somewhat volatile, mirroring the fundamentals of the aluminum industry, we have focused on providing a stable cash return to our investors. The five-year payout ratio has been between 80%-180%, partly reflecting the weak earnings in this period, but also emphasizing the commitment we have to our shareholders for the dividend payments.
All in all, it is clear that we do remain focused on the issues that we can control ourselves while we operate in a volatile and sometimes also a very uncertain market environment. The financials reflect a very prudent as well as conservative financial framework and philosophy. I think we should look into this in somewhat more detail. Clearly, we strive to lift our potential for generating cash flow from operations, by continuously optimizing costs, improving efficiency and productivity, leveraging the synergies that we have between the business areas, high-grading the portfolio to strengthen our margins and market positions. In addition, focusing on optimizing our working capital in order to reduce the cash tied up in our business. We are adamant about maintaining an investment-grade credit rating with a strong balance sheet as well supported by a strong liquidity position.
Again, as we believe this is one of the best tools we have in order to manage the volatility in our industry. We are striving for a disciplined and return-driven capital allocation at all times. This involves maintaining and improving our asset base and to ensure the operational excellence that we have, as well as the safety of our operations. We also grow selectively in order to meet our customer demands, to gain positions in attractive market segments, and to increase the higher value add in our business. In lower parts of the cycle, our strong financial position opens up for M&A optionality for better times. We have the opportunity then to invest in some of our attractive organic growth opportunities. The philosophy is clearly to offer a competitive and reliable dividend to our shareholders.
This we strive to maintain throughout the cycle, as long as we can remain investment grade-rated. We would much rather supplement the payout with special dividends or share buybacks in periods of excessive earnings, than cut the dividends in tougher times. Finally, when it comes to the cyclical nature of our industry, we do believe in being exposed to commodity volatility, allowing us to benefit in full from the upside, while protecting the downside with our strong financial as well as competitive position and our diversified business model. Let's then proceed to look into each of the elements in some more detail. Improvements have really become a Hydro trademark, and it's been a very important tool for us to strengthen our relative industry position and lift our earnings potential by influencing what we can control ourselves in our volatile environment.
If we start with earnings back in 2011 of roughly NOK 6 billion, and this excludes extruded products. Since then, we've experienced significant currency development in our favor, as the dollar has strengthened against some of our main cost currencies, the NOK and the BRL. Now this is a net currency effect, as dollar-denominated costs has also increased, partly offsetting the benefit that we had on the revenue side. We've also seen positive effects from decreasing raw material prices, and this is particularly coming through on the energy side, which is partly offset by some increased costs on the ICMS in Brazil. At the same time, prices for aluminum and alumina have declined more than offsetting the positive currency and raw material effects for the period.
In addition, we've also seen quite some negative effects from inflationary pressures in particular coming out of Brazil. As you see on this slide, if we did not run any improvement programs since 2011, and also for the sake of consistency, we've excluded the improvement programs between 2009 and 2011 in Primary Metal, we will see a close to zero EBIT here today in 2016. Therefore, when we have an annualized EBIT of NOK 5.3 billion, this is almost entirely driven by the NOK 5.6 billion in improvement programs that we have delivered so far. I think it's also important to say that the improvement efforts that we do in Hydro, it's much more than just a primitive cost-cutting exercise.
There are a lot about the way we work on a day-to-day basis, really about stability and control of our operations. On the pie chart to the right here, we show you an approximate split of the B to A ambition by different categories. It does involve volume and capacity increases, such as lifting production above nameplate capacity in BNA, and also about capacity creeps within Primary Metal. We work on fixed cost reductions throughout the company, but also variable and process cost improvements, such as higher raw material efficiencies, which we clearly saw in third quarter within BNA, as well as commercial improvements aimed at lifting margins through, for instance, the portfolio high-grading that we do in Rolled Products, as well as logistical optimization as we do in Bauxite and Alumina.
Through these improvement programs, they have contributed to much, structurally much better cost position and stronger margins on the upstream side of Hydro. Now if we start with Primary Metal, we see a gradual reduction of some 26% in the all-in implied primary cost since 2012, while the realized price fell by some 23%, resulting in a relatively stable margin when you measure this in dollars. On the other hand, if you add to the equation a 45% stronger dollar against the NOK, an 80% stronger dollar against the BRL compared to 2012, we will see more than a 50% improvement in the implied margin when you measure this in Norwegian krone. The picture for BNA is very much the same.
We have a large BRL-denominated cost base, which has become lower when you measure this in dollars due to the significant weaker BRL. The implied alumina cost has gradually come down by some 26% since 2012, while the realized alumina price is down some 20%. This gives us an implied dollar margin almost doubling. If you convert this to Norwegian krone, we've seen an almost tripling in the margin in BNA from 2012. The bottom line of these slides is that we are in a very much stronger position today compared to where we were four years ago, not only driven by the macro factors, but also very much driven by the improvement programs run in the company.
We've also been working hard on strengthening margins in the downstream operations, Rolled Products, as well as in Sapa. Also here, we are very pleased to see good and strong developments. Now if we take Rolled Products first, the underlying EBIT per ton, if you exclude Illinois smelter, has improved by almost 60% from the 2013-14 levels of about 800 NOK per ton to the 1,200-1,300 NOK per ton that we see in 2015 and 2016. This is clearly a result of the ongoing operational improvements that we've done, as well as the portfolio high grading that we've done in an industry which is extremely competitive and also subject to strong margin pressure. In addition, there's also been some positive currency effects also affecting Rolled, in particular in 2015.
If we then look into the underlying EBIT per ton for Rolled Products, if you include Neuss, then we see a somewhat lower margin level over time and a somewhat deterioration over the last 12 months. It is important to note that the results in Neuss, they're still suffering for a what we can call an unfavorable energy contract. This will be replaced by a new energy contract from 2018 and onwards. This will, in effect, reduce the cost in the Neuss smelter by several hundred $ per ton and lead to a substantially improved cost position for Neuss, creating a significant upside for Rolled Products.
If we look at Sapa, we have seen a continuous and quite impressive improvement in the underlying EBIT per ton since the establishment of the joint venture in 2013. This is clearly the result of the restructuring as well as the synergy program completed one year ahead of plan back in 2015. We've continued to do operational improvements. Subsequently, there's been a relentless focus on value over volumes, adding advanced customers to the portfolio, also supported by strong growth in North America. Another way to free up cash is of course to focus on net operating capital. If we look at the four quarters rolling net operating capital in terms of days, and compare it to 2012, we see a reduction of 7 days, representing roughly NOK 1.5 billion in freed-up cash.
This, of course, is partly driven by lower prices, but also a real optimization of stock levels as well as train terms during or throughout the value chain. However, if you compare this to 2014, we can see that we've increased working capital by roughly 7 days. This is mainly due to the above average buildup of working capital between the business areas that we all saw in the beginning of 2015. We are now finally on the downward trend again, following good progress of unwinding this inventory buildup and then getting more towards what we can call a normal level. However, there's still some way to go, and we will continue to focus pretty sharply on this coming into also 2017.
So far this year, we've had a normal seasonal buildup as in working capital since the end of 2015. We do expect this to be partly released on the back of normal and seasonal destocking activity towards the end of the year. A continuous portfolio review and divestments is also a source of cash flow for our company. Over the last 3 years, we've divested several, mainly smaller assets, but we also still have a few ongoing projects in the pipeline. As a large company with a long history, Hydro also has a number of legacy assets in the portfolio, which we are working to clean up. Some of those actually may have a notable cash flow effect, like the Herøya Industrial Park that we divested in the second quarter this year.
While others reduce future environmental obligations and the risk related to those. In addition to the legacy assets, we are continuously reviewing our non-core assets that are not performing as well as we hoped. A good example of this is the divestment of the Slim rolling mill that we did in 2015, which had a positive effect both on the Rolled Products results, as well as the average margin in Rolled. When we talk about balance sheet strength, we first and foremost talk about the importance for us of maintaining investment-grade credit rating. This of course allows us to access the bond market when we need to, and to get competitive terms. It also enables us to realize business opportunities where there is a need for meeting a requirement to be a very credit-worthy partner.
Since the start of 2011, we've held a BBB credit rating, despite the fairly weak market prices and aluminum sentiment that's been in this period. In 2016, Moody's basically placed the entire metals and mining sector on negative watch, including Hydro, resulting in quite a few downgrades in our industry. I'm very happy to reiterate that Hydro came through this, continuing to carry our BBB rating. To make sure that we satisfy the criteria of the rating agencies, there are two key targets that we try to follow over the cycle. One target is to have adjusted net debt to equity below 55%. The second, funds from operations to adjusted net debt, should be above 40%.
The first ratio, adjusted net debt to equity, has been comfortably above the target since 2009. The other ratio, however, funds from operations to adjusted net debt, has been quite a lot more challenging, reflecting really the weak pricing environment and the corresponding earnings in Hydro. As such, the funds from operations has really been a restrictive factor when we have determined what is an optimal balance sheet structure for Hydro over the last couple of years. As you can see on this slide, it is really only in the last two periods on the back of improved earnings that this ratio has been well above target. It was 89% back in 2015, and it's been above 85% for the last four quarters.
Let me on this slide just also quickly comment on the liquidity position, as this is an important source of financial flexibility for the company. As communicated in third quarter, we have roughly NOK 8 billion in cash and cash equivalents. In addition to that, we carry NOK 4.4 billion in what is called short-term investments, which in reality is cash. It just takes a little bit longer time to realize it. In addition to this, we also have $1.7 billion in a revolving credit facility which matures in 2020. Let's now briefly look at how we define adjusted net debt, as this is used as the basis for the 2 key targets that we just discussed. During the last 4 quarters, we've been able to reduce adjusted net debt by 18% or roughly NOK 3 billion.
The main reasons for this include a NOK 2 billion higher net cash position of NOK 5.4 billion at the end of the third quarter. This is primarily driven by cash flow from operations. We have a substantial reduction in net debt in Qatalum as well as in Sapa. This has been reduced around NOK 2 billion. This has been driven primarily as a result of positive cash flow in these two entities, and then a small part is the result of a stronger dollar versus NOK, leading to a balance sheet reevaluation. Operating lease commitments and other adjustments have also been reduced with NOK 600 million. This is mainly due to the changes relating to the operating lease for the Alunorte bauxite vessel, as well as changes around the Herøya Industrial Park offices.
Please keep in mind that the adjustment for the Alunorte vessel has just been moved from as an operating lease into long-term debt, so the net effect is zero as such. These positive effects partly offset by NOK 1.5 billion negative adjustments on pension obligations, primarily driven by lower interest rates environments in both Norway as well as in Germany. We turn to CapEx and we start with sustaining capital. Sustaining capital, of course, is needed to keep assets in good shape, but it does also include investments that will lead to better performance, better efficiency, and also improved safety of the current asset base. As some of you may remember, last year we did establish a long-term sustaining capital level of NOK 3.5-NOK 4 billion per year.
This year, we have performed another evaluation of the CapEx profile, and in order to be on the conservative side, we do feel it is prudent to lift the range up to around NOK 4 billion per year. Now the main reason for this is the number of large sustaining projects and investments in Brazil, combined with also high inflationary pressures in Brazil as such. Let me also just point out, for the sake of good order, that NOK 4 billion is still well below the annual depreciation of NOK 5 billion for the company. Now if you first look at the historical figures for 2011 to 2015, you will see that we've been able to maintain the sustaining capital at relatively low levels below NOK 3.5 billion on average.
If we then include the expectations for the next four years, the average for this entire nine-year period would be around NOK 4 billion. However, the level for the next four years is elevated and approaching roughly NOK 5 billion on average. The main reason for the higher than average CapEx for the next four years is primarily driven by the large sustaining projects in Brazil shown on the light blue color on the slide. Most of you are familiar with the two long life project, the new tailing dam in Paragominas, as well as the new red mud deposit area in Alunorte. In addition, this year we also have to include the opening of the new mining area in Paragominas. Last year that was primarily coming in 2019 and beyond.
We've had to put that, pull that forward a couple of years given the high production level that we've been able successfully to do in Paragominas. If we then look at the expectations of, for 2016, this is now NOK 5.3 billion, which is somewhat below the guidance of NOK 5.7 billion, mainly as a result of CapEx optimization within the primary metal business area. If you look at sustaining CapEx for 2016 and 2017 together, that is roughly the same as we guided on last year. We then complete the CapEx picture with the ongoing growth projects.
Even though that we in the current market environment, we really do not see room for any large organic growth projects, we do see opportunities for profitable growth in certain selective areas, such as recycling, portfolio high-grading, as well as developing technology in order to remain competitive in terms of costs, market shares, and market positions. We've already completed several projects supporting our strategy here, such as the UBC recycling facility in Rolled, the Clervaux recycling upgrade in metal markets, the Alunorf debottlenecking in Rolled, as well as the adjustable flexible mold technology as we've seen in primary metal. We still have a couple of large projects still ongoing. Still in the finalization phase of the new automotive line number three in Germany.
Importantly, the Alunorte debottlenecking, bringing it up to an annual capacity of some 6.6 million tons. On top of this, we also have the large investment in the Karmøy technology pilot that was approved earlier this year. We expect to be some 70% completed by the end of 2016. The net investment amount of NOK 2.7 billion is still unchanged, including the NOK 1.6 billion that we have received in support from Enova. One of our key priorities is of course to contribute to running Hydro with a strong and sustainable long-term value creation reflecting in the competitive shareholder return, total shareholder return.
As TSR is affected by the industry and market fundamentals, and they will be volatile, the philosophy that we have as a company is to have a stable and reliable dividend that is sustainable throughout the cycle. Again, as long as we can satisfy our balance sheet requirements, as we have discussed before. This philosophy is reflected in the dividend policy, which remains unchanged from the last update in 2014. The dividend was then set at 1 NOK per share, up from 0.75 NOK per share, and should really be considered as a floor when looking forward. At the same time, the payout ratio was increased from 30%-40% of net income over the cycle.
However, in the low earnings environment that we've seen in the last five years, the payout ratio has been closer to 110%, indicating that in some years we've paid out somewhat less, and in some years, we've paid out quite a lot more than what we theoretically could afford. As Svein mentioned, according to normal procedures, the board will review the dividend policy and dividend level in connection with the fourth quarter results in February, and put their proposal forward to the general meeting to be held in May. As we operate in a cyclical commodity industry with a significant exposure to commodity prices and currencies, having a thorough risk management policy in place is crucial. Hydro's philosophy is to remain exposed to market fluctuations for the most part.
As such, we do very limited or no financial hedging, which exposes us to the downside in lower cycle periods, but allows us also to benefit in full from the commodity cycle without capping the upside. Our strong conviction is that the best tools at our disposal is to manage volatility with a prudent financial policy as well as the flexibility embedded in our integrated business model. First and foremost, a strong balance sheet is the best tool against cyclicality, as we have discussed before. Continue working on strengthening the competitive position is another way, and one of the best ways to navigate through tougher times. Then finally, running a diversified business model allows us to give good performance, as volatile earnings from the upstream side is somewhat offset by stable margins in our downstream divisions, Sapa and Rolled Products, as well as from the energy business.
We predominantly maintain our long-term debt in dollars, which is the primary revenue currency, again, to structurally mitigate exposure risk. As a producer company, we do some operational hedging on a continuous basis to match the customers' buying patterns as well as securing our margins. Before we move on, let me just give a few words on explaining why we do not believe in extensive financial hedging. Now, historical evidence suggests that there are significant cross-correlations between exposure elements over time, as illustrated by this correlation triangle. Aluminum and oil prices have shown a strong negative correlation with the US dollar as the commodity currency. At the same time, the oil price is closely linked with the NOK due to Norway's high exposure to oil.
Finally, the oil price is positively correlated with the other commodities, both due to sentiment as well as they typically are part of the cost base of the other raw materials that we use. Now, these cross correlations form an imperfect, yet quite notable natural earnings hedge for Hydro, as I've also illustrated earlier on the implied cost slide. As such, if we hedge one leg of the exposure, we may automatically increase the risk rather than ending up with a lower risk. Now that we looked into the correlations between the main exposure elements, let's turn to Hydro sensitivities in different commodity price scenarios and currency scenarios and discuss earnings and returns. We focus on the LME and the dollar as they do have the largest impact on our earnings.
A 10% increase in the LME base of $1,625, which is what we realized in the third quarter this year, would add some NOK 2.9 billion to Hydro's underlying EBIT on an annualized basis. This is the net effect of both the revenue side and the costs linked to LME. The majority of this 2.5 billion comes out of the LME exposure that we carry in Primary Metal, while the LME exposure in BNA is declining due to the lower share of LME-linked alumina contracts. Changes in the US dollar against the Hydro currencies, the primary Hydro currency NOK, euro, and BRL also have a significant effect on our earnings. As most of our commodities are quoted in dollars, while the majority of the fixed costs and some of the variable costs are in local currencies.
The underlying EBIT would improve with some NOK 2.8 billion, given a 10% strengthening of the dollar, almost the same effect as the LME change. The sensitivities on the financial items mainly reflect the one-off unrealized revaluation effects on either debt, intercompany positions, or embedded derivatives in the power contracts that we do have. The euro is quite clearly the largest exposure on this side. Now, if we look at the sensitivities in practice, let me first elaborate more on the topic of cost sensitivities between LME and the dollar NOK exchange rate. The LME sensitivity would, of course, clearly be affected by the level of dollar NOK, and the dollar NOK sensitivity would also clearly be affected by the LME level.
This effect becomes quite visible when you do large simultaneous movements in both variables, which I would like to illustrate with a sensitivity matrix. The starting points are the sensitivities that I've just showed you, with EBIT moving NOK 2.8 billion on a 10% change of the dollar NOK, and NOK 2.9 billion on the 10% change in LME. If you look at the corners of the matrix, we see that the combined effect of a simultaneous change in LME and dollar NOK is quite different than just the simple sum of the sensitivities due to the cross effects. In addition, over time, please remember that we're more likely to see the variables moving in the opposite directions, again, pointing to the natural earnings hedge. Although in shorter intervals, they can move in the same direction.
Let me then also take the opportunity to apply the sensitivities in a very simple run rate model, demonstrating the effect of the latest developments in the key variables on Hydro's underlying EBIT and EPS. We've used spot rates compared to the levels that we realized in third quarter, indicating an upside in our earnings with higher LME, higher PAX, stronger dollar versus NOK, and a weaker BRL versus the NOK. Using these sensitivities, we see that this will have a positive underlying annual effect of around NOK 5 billion on EBIT or 1.9 per EPS per share. Let me then summarize my financial update with a few scenarios, and then we start with the EBITDA scenario.
Please keep in mind, as we have said now several years in a row, that these are not estimates for the coming years, but they are indicative ranges using earnings for the last four quarters as the basis, and then applying the sensitivities that we've just been through. This is an illustration of the relative changes in our earnings with and without improvement efforts on the different LME scenarios. Please also remember that these scenarios are simplified, and there are several positive and negative additional factors that could influence our earnings that are not taken into consideration here. We will look at three pricing scenarios. The middle scenario with LME of $1,700, USD/NOK of 8.5, BRL/NOK of 2.5, aluminum price of around $300, all relatively close to the spot rate.
We've also assumed a Hydro realized premium of $275 per ton, which is close to what we achieved in the last four quarters, and is also within the range that we've guided for the fourth quarter. By keeping all these variables constants, we then create a low price scenario with LME at $1,500 per ton and a high price scenario at $1,900 per ton. These are not price forecasts, but only reference points indicating a downside spot case as well as an upside from where we are today. With these disclosures, let's look at the two different EBITDA ranges. First, as is, including the NOK 5.6 billion in realized improvements since 2009. With the lowest price assumption, we get to an EBITDA of around NOK 11 billion.
In the spot scenario, around NOK 14 billion. With LME at $1,900, we would get to approximately NOK 18 billion in EBITDA. If we then add on the remaining improvements of NOK 1.9 billion until 2019, the EBITDA earnings would be lifted to roughly NOK 13 billion in the low case, NOK 16 billion in spot, and roughly NOK 20 billion in the higher price range. Again, clearly demonstrating the value of the Hydro improvement efforts. If we then look at some ROACE scenarios with the same price ranges and under the same assumptions as before. Also, the disclosures I made on the previous page is also relevant for these scenarios. Svein has already shown you the impact of the realized and potential improvement ambitions at spot or in the middle price scenario here.
Let me also comment on the down and upside cases. If we again start with the blue as is scenario, the ROACE with the already realized improvements would be around 5% in a low price scenario, roughly 8% at spot, and approximately 12% on the upside. If we realize the full potential of the remaining improvement ambitions, the ROACE in green would be around 7%, 10% and 13% respectively in the low spot and high price scenarios. Again, you can see that with the improvement efforts together with a little help from the market, we would have a sustainable ROACE level over time. If you move to the last chapter of my presentation and comment on Hydro's financial targets as well as aspirations.
I cannot emphasize it enough that in our volatile industry, having the financial muscle and flexibility is crucial for being able to navigate through the cycles. As such, one of our top priorities, especially in a down cycle, is really to invest in our own balance sheet. A strong balance sheet does protect us against the actions of last resort, such as fire sales of assets or dividend cuts, and allows us to concentrate on performance as well as pursuing attractive opportunities which may appear as other companies are struggling. Now, as we operate in a capital-intensive business, CapEx is necessary in order to maintain and improve our position in a fast-paced and competitive environment.
These include maintaining existing assets as a prerequisite for lean and safe operations, but they also include selective investments in projects allowing us to keep our market share, strengthening our margins, as well as improving our costs. The next priority would then be to deliver reliable and predictable dividends to shareholders, regardless of where we are in the cycle, as long as we maintain our balance sheet requirements. As the commodity cycle improves, so will our funds from operations. More options for capital allocation becomes viable. Over a longer term, we do aim to create shareholder value through growth by reinvesting excess cash flow in organic growth projects or pursuing attractive M&A opportunities, for which will of course satisfy our over the cycle return requirements. In the absence of profitable growth opportunities, we do aim to return excess cash to our shareholders.
To end my presentation, let me then summarize our key financial targets and ambitions and update you on the results so far. First, delivering on the NOK 2.9 billion better improvements for in the period of 2016 to 2019 is, of course, a key target for us. We have already delivered NOK 1.1 billion in 2016, very much in line with the guidance for this year, and we remain confident that we will deliver the rest of the ambition. As we have discussed, we believe it is prudent to adjust the long-term sustaining CapEx guidance upwards to around NOK 4 billion from the previous level that we guided on of NOK 3.5-NOK 4 billion. This is done on the basis of inflationary pressures in Brazil as well as a portfolio update.
If we then look at the total CapEx, if you include growth, the average level for 2017-2019 is expected to come in around NOK 6 billion per year, including the net investment in the Karmøy Technology Pilot. In 2016, we do expect NOK 7.8 billion in total CapEx, which is slightly below the NOK 8.1 billion we expected earlier, mainly due to the lower sustaining capital in 2016. In 2017, we guide for CapEx reduction to NOK 6.8 billion. The payout policy is 40% of net income over the cycle, but over the last year, the payout ratio has been 110%. As I've said before, we should also here think about the NOK 1 per share as a floor of dividend that we do intend to maintain.
As we've discussed, the two important financial targets that we use to evaluate the strength of our balance sheet that provide a framework for capital allocation, they are funds from operations to net adjusted debt has to be above 40%. Over the last four strong quarters, we have seen a ratio of 85%, well above the targeted level. The second target, funds from operations to net adjusted debt. Second target, adjusted net debt to equity should be below 55%. This was 17% at the end of third quarter, also comfortably within the target. I'll rephrase my first point, which was funds from operations to net adjusted debt should be above 40%, and we've seen 85% in the last four quarters.
Finally, if we look at ROACE as a measure of sustainable profitability and value creation, then our ambition is of course to deliver a ROACE well above cost of capital over a cycle. Also at the competitive level compared to our peers. In the last four quarters, we've seen a ROACE level of 5.4%, down from last year and below the level that we will believe is sustainable for Hydro in the long term. To conclude, in current markets with uncertain outlooks, one of our key priorities is, as I've said, to remain financially agile, to find balance between capital distribution and financial flexibility, to maximize long-term value creation potential in today's environment.
We are focusing on strengthening the relative industry position through the improvement programs that we do, as well as investing in our own balance sheet, enabling us to maintain a safe and predictable cash return to our shareholders. We do strive to further optimize our working capital as well as for disciplined and return-driven capital allocation regardless of where we are in the cycle. The effective risk management allows us to navigate through the stormy waters and do act on attractive opportunities if they should arise, aiming to drive long-term shareholder value creation. Thank you.
Thank you, Eivind. Could Svein Richard Brandtzæg please rejoin us on stage? We are then ready for Q&A. There will be two microphones on each side here, and please state your name and company when asking questions. Menno?
Good morning. It's Menno at Morgan Stanley. Two questions, please. First on CapEx. This is the second year in a row, maybe therefore correlated with being in London, that you've done a review of your portfolio and on the CapEx, and again, things went a bit higher. Can you talk us through what the drivers exactly are? You mentioned inflation a couple of times, but you know, you could see that last year clearly that inflation was coming. Is there something wrong with the assets? Is the engineering of the projects bad? What's driving it? And the second one, you talked about sustainable profitability levels and not having been there for a while. Can you frame for us what, in your view, a sustainable profitability level for the industry is?
Do you think other companies share your view of a sustainable profitability level? Because if they don't and they invest despite that, who cares what your views are?
If I do the sustaining capital one?
Yes.
I think one of the primary drivers, Menno, on the sustaining capital has also been to bring the expansion in the Paragominas mine up with a couple of years. It's a new area which we call M5 or Miltônia 5. What you didn't see last year was that it was planned to do a startup beyond 2019. This year we have to estimate that we have to start that work already in 2018. Now this comes very much on the back of the high performance we've had in the Paragominas mine, and you look at the high production profile that we've had. That's a significant part of this.
Okay, we've got to sustainable profitability. I think, in general, that should be above cost of capital, of course. I think our competitors are using the same calculators as we are doing. I think they also realize that, they've also been participating in an industry that has not been able to deliver sustainable returns. I don't believe that there was somebody that or believing that it has been a satisfactory profitability level in this industry for quite a while, because that is the reality here. This is also the reason why we are quite motivated to continue our improvement programs.
Including the Chinese?
In China, I wouldn't talk for the Chinese here, but I think also when we look at the situation in China now, it's also on the way to be changed. I think there could also be more realistic say adjustments towards sustainable profitability also in China.
Eivind Rygg, DNB Markets. A question on maybe longer term strategy on recycling has a stronger growth than primary production or primary demand. I just wanted to ask you how you want to leverage on that longer term. Could we at some point see production from recycling at the same level in your portfolio as actual electrolysis? Or would you continue to grow in primary production? Thank you.
Thank you very much for a good question. I'm sure that Kjetil Eidesmo will really be happy if he could have a recycling strategy where the volumes would come up to the same level as primary production, but that will take some time. On the other hand, as I talked about, recycling is a important area for us and our strategy is to increase recycling, and we have the target of 250,000 tons by 2020, and we are absolutely moving in that direction.
Hans-Erik Jacobsen, Swedbank. You mentioned that in the longer term an M&A may become a possibility again. In which areas of your different businesses is that most likely? In other words, where would you like to grow?
Well, as I said, we have a long value chain with opportunities. With a strong balance sheet, we can also do counter-cyclical investments. It depends very much on the right timing or the right asset. We are not going to invest in low quality carbon-based primary production. I don't think Hilde will like to run these smelters. There are certain limitations there, but also, I would say, opportunities along the whole value chain that we see that could be of interest for us. Again, the basis for us is not to be the biggest in the world, it's really to be the most profitable company in the world.
That again will be quite targeted opportunities that we are looking at that can create higher shareholder value, higher profitability for our company. I wouldn't rule out in a way any investments along the value chain here. We see there are some opportunities and we are following it carefully.
Thank you. It's Danielle Chigumira from UBS. A couple of questions. Firstly, just going back to CapEx. If I compare your current CapEx guidance with your third quarter, it seems like the increase has been rather on the growth CapEx rather than the sustaining, because you've explained the sustaining already. Can you give us some more color around where you're spending that growth CapEx over 2017 and 2018, excluding Karmøy, where we've got pretty good visibility on that? And the second question is just on the free cash flow scenarios you gave. You just flexed the prices, which is understandable why. Could you give us an idea of at spot aluminum prices, how much would the NOK have to strengthen for you to get to free cash flow breakeven?
You take that one.
If we start with the CapEx side, as you said, I mean, there's a lot in there on Karmøy Technology Pilot. There's still capital in there for the automotive body-in-white project in Germany. There is also quite some growth CapEx in the number related to the different creep projects that we plan to do within Primary Metal. In order for them also to reach the additional 100,000 tons that we plan to do on this side of 2020. That's in a way the part of the equation that you don't see on the slide. On the second one, Danielle, let me look at my numbers and get back to you on that.
I don't take that in my head, on my feet.
Hi, Fraser Jamieson here, J.P. Morgan up here. Couple of questions. Firstly, on MRN or the increasing stake in equity bauxite. You've got that as a red dot in your kind of strategic priorities diagram, if you like. Is that deal now completely off the table? You obviously said at the time that it was a question of commercial terms. Is the fact that that's a red dot now saying that it's no longer a strategic priority at all for you, or is it potentially something that could come back? The second question, sorry to go back onto CapEx yet again.
Following up on both Manuel and Danielle's questions, there has been a pretty significant move, you know, in a matter of a few weeks really since couple of months since you published your third quarter numbers, when we had the previous profile out there. Could you give us a sense of what has happened in terms of planning cycle, et cetera? Is this to do with budgeting processes that have come through over the last few weeks?
Okay, I can start with MRN and, as we have communicated previously, we didn't simply agree on the price with Vale. If Vale comes back again to the table, there could be a new negotiation, but we didn't agree on the terms there. That is really the reason for it. We are not taking it away from the target, but it's a red light on the equity book side due to that situation.
On the CapEx side, if you remember third quarter, what we see today is I guided also in third quarter that there is a risk or possibility that we will spend less capital in 2016 than what we had guided for so far. That we now see coming through in the numbers I've presented. This is what we typically in Hydro call CapEx optimization. If we have projects that we can take from 2016 and move them into 2017, we will do them simply from a net present value perspective. That is part of the reason why we see a little bit of an increase also in 2017.
If you average those two years out, it's pretty much the same as we guided for at capital markets day in 2015. The other really new part into the longer term sustaining capital is really what I've answered also Manuel. Has to do with opening up a new mining area in Paragominas on the back of the very strong production performance we've had in that mine.
Okay, it's Johannes Grønseth here at Handelsbanken. I have a question on investments in rolled products and what kind of opportunities do you see there for converting more low margin rolled products into the attractive OEM segment? Anything there? And if you... I understand you're only active here in Europe. Do you foresee any possibilities there in other markets?
That's a good question. When we look at order intake for Automotive Line 3, for example, this is growing faster than what we expected when we decided on this investment. Meaning that there are more customers knocking on the door, they would have more volumes. We see that there are so many OEMs now that are converting from steel to aluminum that they are concerned about if there is enough capacity, although we feel that there is enough capacity short term, but longer term, there could be more opportunities. Of course, we are number two in Europe now with the position we are taking.
We are going to take the time now to ramp it up and stabilize the production at full speed, and then we'll see what can come next. This is an interesting market. With regard to operations outside Europe, that has not been decided, but we are selling products outside Europe. We are producing body in white material in Germany and exporting to key customers in the North American market already today.
Do you have any plans outside Europe which is suitable for converting or making these kind of brownfield investments in?
We don't have any concrete plans about that.
Eirik Melle, Danske Bank. I have a question about your position on the cash cost curves. If you look, I guess you're moving a bit up and down depending on currencies, as you mentioned. If you look long-term, if you look at, you know, trends on capacity coming on stream, going out in China, but also in the world outside of China, and your peers, what they're doing on costs and improvements, what do you see as a direction in the longer term for the company as a whole?
The position on the cost curve depends on several factors, but competitive energy is absolutely necessary. If you don't have competitive energy contracts or energy sources, then you have a big problem to be in a low part of the cost curve. Operational excellence, to be able to run the smelters with low energy consumption, with low cost on the input factors, and again, with the parameters that is leading in the industry is also very important if you are going to look for positions down on the cost curve. Hydro is very well positioned in that respect.
We see that where we have the majority of our smelters, for example, in Nordic power market, we see that the energy cost has come down quite substantially during the last years. Also looking forward, it's quite competitive environment for aluminum production going forward. Of course, this is another competitive factor coming up due to the fact that we have renewable energy here. With coal prices coming up in China, that could be temporary, but we still believe that we are quite well-positioned in the aluminum industry with regard to energy.
Also with regard to operational excellence and also utilizing the technology, due to the fact that we have a leading position in technology in the global aluminum industry, nobody is even close to what we are doing with regard to energy consumption for aluminum production. We have a big advantage. Of course, we will build on these advantages also going forward.
We have time for one more question.
Boris, IG. Could you comment on your stake in Sapa going forward? Is this a long-term investment?
Yes. When we established Sapa as a joint venture in 2013, it was for us very much to take the opportunity to do the restructuring that was necessary in the extrusion industry in a better way seen from industrial and market point of view, together with the Sapa extrusion businesses, and to create an option for Hydro. The option for us is still valid. We know that at a certain point of time in the future, Alcoa is going to leave, and then we could decide to leave together with Alcoa and make an IPO. Or we could negotiate a price for their shares, acquire their shares, or we could end up in something between that we have a minority position in Sapa long term.
This option is still valid today, and that is the situation that we have maintained over the years since we established the joint venture.
Thank you. It's time to call for the first break of today. We will have a short break, and please be back here at 10:05 A.M.
10:05.
Senior Vice President of Corporate Strategy and Analysis, Kathrine Fog.
Thank you, Stian. Good morning, everyone. I would, I'm happy to take you through the market outlook this morning. We'll go through the macro outlook and downstream. Secondly, I'll look more into the primary metal section, and then we will go through bauxite and alumina. And then to close, we will look at our long-term market outlook. Let's start by macro and downstream. Now, the global macro development is improving, but there are some uncertainties. The macro outlook remains moderately positive, driven primarily by emerging economies, China, India, as well as other Asia. The Eurozone and the US, we expect to be moderately positive, kind of steady. Of course, over the last few weeks, we've seen some added uncertainty in the US following the presidential election. China's growth is in line with government targets.
We expect it to continue to be in line with government targets, focusing on stability, especially towards next year's Communist Party Congress in November of next year. It seems as if stability is an important factor, the name of the game. India is showing strong growth, but will not, in volume terms, rival China. Of course, as we all know, macro is a main driver for aluminum demand, as we will return to throughout my presentation. Global aluminum demand prospects remain encouraging. We, as you see here on my left-hand side, and as we talked about before, aluminum is well-placed to enjoy demand across the different phases of the economy. With a diversified consumption base, aluminum is insulated as the economy moves from being investment driven towards consumption driven.
It grows with building and construction segment, infrastructure investments, but it also grows in the consumption sector with packaging, with electrical and household investments as the economies become more urbanized and consumption driven. In the mid slide, we see also as in previous years, and as we heard before also through the CEO presentation, automotive remains one of the main driver for growth. On the right-hand side, looking further ahead, we also see that global aluminum is increasing, both with population growth and also as aluminum intensity continues to increase across the economy. As I said, and it was also mentioned by Svein Richard Brandtzæg, we see strong substitution trends in continuing in automotive.
Now, we talk a lot about body in white, but we wanted to share the whole car with you, as you see on the left-hand side. The automotive segment is driving aluminum growth through two main trends, both through GDP growth and increased GDP per capita, especially in emerging countries. It's also driven by growth in aluminum per vehicle, so driven by substitution. Through the continued need for lightweighting, we believe that aluminum in cars will grow by some 30 kilos per car for the period 2015 to 2020. Now, as you see in the slide here, this is happening throughout the different applications into the car. It's in body sheet, hang-on parts, typically in the rolled segment.
That's where we see the highest growth, around some 7% CAGR for this period. We also see the growth in extrusions. That's typically in the structures of the car, in seats or the frames. We also see it in castings. That's the lower blue area here, which actually constitutes the largest share of aluminum going into the car, typically engine blocks, powertrains, steering components. The two latter ones are growing with the number of cars, while the first one is growing also with substitution. Looking at the rolled segment more overall. Yes, as in previous years, demand in the rolled segment is driven primarily by automotive and other transport for the reasons that I've just mentioned.
We also see decent growth in other areas, typically the packaging segment, which is of course important due to its size in the Rolled area. This is especially in consumer packaging, main drivers being urbanization, but also some substitution. Rolled Products demand is less cyclical, has a weaker link to GDP growth than what we see, for example, for extrusions. Demand in the rolled segment is growing with general consumption patterns and will be less volatile through the GDP cycle. Following somewhat lower demand in 2016, especially in the U.S., our expectations for 2017 is a return to growth levels around 4% in the U.S. and around 3% in Europe. We also expect continued growth in extrusions.
Now, the extrusions demand is primarily going into the construction sector. As we see here, around 60% or more than 60% going into building and construction. Other important segments, transport and machinery. In the US, we have seen strong growth in the building sector over several years. Now we do expect the growth levels to moderate somewhat, but we've also seen quite recently indicators showing good momentum also into 2017. In Europe, we expect somewhat gradual improvement. We're seeing some growth again, but from low levels in Southern Europe, while northwestern Europe is expected to remain relatively stable. For both markets, we expect growth around 2% in 2017. Now, also elsewhere outside these regions, extrusions is growing primarily with building and construction and infrastructure investments.
Now, the extrusion market is, as I said, supported by continued momentum in building and construction. 40% of global energy is going into buildings. Therefore, to improve energy efficiency and reduce the carbon footprint of buildings is getting increased attention. Europe and U.S. are in the driving seat in this area. Aluminum-based solutions are important contributors to meet climate requirements for new buildings. We see this especially in the commercial sector. Typical products are window frames, it's curtain walls, facade systems, integrated systems. As a consequence, demand for aluminum into the building and construction segment is growing both with general growth in building and construction. We see typically in the U.S. increased single family unit growth. The segment will also benefit from an emerging drive towards what we call green buildings through regulations, increased focus on green public procurement, but also from customer demand.
We also see this in legislation. We have the EU Energy Efficiency Directive from 2012. Now, yesterday, some of you may be aware, the new Winter Package, came out or proposal came out in the EU, where the energy efficiency target was strengthened further from 27%-30% for 2030. This will mean quite a bit also in the building sector. We also see in the state of Texas, decided earlier this year to apply the International Energy Conservation Code being effective of first September this year, meaning increased focus on the energy efficiency of buildings and of particular solutions in the buildings. We also have the nearly zero energy building concept in the EU. Again, target is directive and targeted for increasing energy efficiency of buildings and targeting specific solutions.
Let's move over and have a look at China. We see China's consumption pattern for aluminum changing somewhat, with transport emerging as a key engine for growth. Chinese aluminum demand has been driven and continues to be driven by building and construction, infrastructure investments to a large degree. We see the spectacular growth in the bars here from 2005 to 2015, the green area, how building and construction has increased. This is partly a result of government stimulus, and especially maybe this year and last have been a result of government stimulus. We don't expect this to continue perpetually. We are seeing that the inventories of buildings outside the most vibrant Tier One cities is still quite high.
With an expected slowing down in the housing sector, transportation is expected to emerge as the largest engine for aluminium growth going forward in China, estimated growth of some 7%-8% for the period 2015-2020. Car ownership in China is increasing. When we were at the recent Antaike Conference in Nanning in Guangxi, we were even a bit surprised by hearing from a lot of the Chinese participants in the meeting talking about the need for and the drive for a medium-sized aluminium car in China for the Chinese market. With an increased focus also on electric vehicles, aluminium seems well-placed to benefit from development in this sector going forward in the Chinese market.
Now what we also see in China, there is a high use of aluminum into other sectors, other parts of the transportation sector, especially into railway and trains, but also into buses and trucks. We also expect aluminum growth to be strong across many of the other segments, such as electrical and packaging. Growth in Chinese semis exports has been largely stable over the last year. The Chinese aluminum production, as we see on the right-hand side here, the green area, is directed primarily at the domestic market. Now, around 90% of the Chinese production is consumed in the domestic market. I can't rule out that some of this is coming out as products containing aluminum, but still. Now the remaining 10%-12%, the blue area here, is being exported.
As illustrated in the graph, the share of semis exports to total production has stayed relatively stable over the last years. However, with an increasing underlying volume, production, of course the volume exported is increasing somewhat. Now if you look at the left-hand side of the graph, the gray area shows monthly development in net semis exports from China. Again, we see that there is an increasing trend. It's been relatively stable for 2016. The two shaded bars here show periods of particularly high positive price arbitrage between China and rest of the world. During these periods, we have seen quite high levels of what we consider opportunistic exports of semis from China, then breaking the underlying trend. This arbitrage window has been relatively stable throughout 2016. Now trade regulations and duties are also impacting trade flows.
In addition to the price arbitrage, trade is affected by regulations, by duties, as well as anti-dumping initiatives. Primary exports from China are subject to an export tax of 15%. Semis exports from China enjoy a VAT rebate of some 13%-15%. You see in the pie chart here, again, that Chinese semis exports are going primarily to the Asian market, around 50% ending up in Asia. While semis exports to Europe and the U.S. have remained relatively stable over several years, we have seen maybe some increase in the exports of flat rolled products, primarily to the U.S. Imports into the U.S. and the EU are also affected by by tariffs going into these regions, both on primary as well as on semis.
We're also seeing that as Chinese semis exports have increased, there has been an increase in the request for anti-dumping measures on imports of Chinese origin. We've seen this for extrusions some years back into the U.S. with an anti-dumping tariff ranging from some 30-some% to several hundred% depending on the product. We've seen it for wheels in the E.U., and we've also seen recently, early part of this year, a request in India for an increase in the general tariff, both on primary and on semis of 2.5%. Now, that is still being debated in India. One may say that it's a fair bet that we may be facing increasing protectionism going forward.
We've all followed the presidential election and the aftermath in the US. The future of TPP and TTIP, so the two large trade agreements that the US are a part of, are uncertain at best. Now, the latest is that TPP will be abandoned during the first 100, not years, but days of the presidency. However, we have also seen that CETA, which is the free trade agreement between Canada and the EU, was recently approved and is expected to be implemented in 2017. We do expect that there will be continued high political attention on trade topics going forward. Let me move on to primary metal markets.
Now first, I'd like to recap a little bit what we said last year at the Capital Markets Day, where we expected the global surplus to moderate somewhat in 2016. We expected demand in the area of 4%-5% and supply in the area of 2%-4%, and we expected a diminishing metal surplus. Now looking at 2016 in retrospect, we've seen that the market balance is progressing actually better than we expected. We've seen demand growth coming in at the higher end of our range, around 5%, while supply has been pretty much bull's-eye of our expectation, around 3%. The surplus is therefore likely to transform into a small deficit by the end of this year, as also shown by Svein Richard Brandtzæg.
We also see that both supply and demand drivers have developed largely as we expected. Now, I'd like to mention in particular that we were pointing to large volume curtailments needed last year. We saw that happening late part of 2015 and into 2016, but we were also looking at new capacity coming on stream. Now we're also seeing that, but we're seeing it relatively late into 2016, the effect being mostly felt probably next year. We have seen demand being more resilient, both in China and in other regions. Now, where will that take us into 2017? Well, we do expect the global market, primary market to be largely balanced also in 2017. We expect demand to grow by 3%-5%. Supply is expected to grow somewhat above this, 4%-6%.
This is likely to give us a market that is relatively balanced to somewhat long in 2017. Of course, this is based on some assumptions, and there are some uncertainties. Now I mentioned that there's significant new capacity coming in at the later part of this year. Now, that's impact will be largely felt into 2017. This is primarily coming on in China, so yeah, in the area of 3 million tons. Also somewhat outside China, primarily in India, some 500,000-700,000 tons. The larger uncertainty is probably the rate of further restarts in Chinese curtailed capacity. At the same time, we do assume that there will be further curtailments throughout 2017, which are needed in order to allow for the new capacity to come into the market.
There are also uncertainties around the cost development. I'll come back to that a little bit, primarily on coal power, but also alumina, and especially again in China as the most important driver. Demand development has some uncertainties as well. The positive outlay, of course, is aluminum into the automotive sector as well as other transportation. Now, if you look at the global balances in some more detail, we see divergence in market balances continuing into 2017. The global balance is built up of a metal surplus in China and a metal deficit outside China. We expect demand growth outside China of some 2%-4% in 2017, while supply growth is expected to be in the area of 2%-3%.
On the other hand, we see capacity expansions in China in the area of 7%-8%, while demand growth is expected to be somewhat slowing down and expected around 4%-6%, and we see this on the left-hand side of my graph here. As illustrated on the right-hand side, when we see this together, our forecast is for a fairly balanced global market into 2017. Again, however, dependent on further curtailments in Chinese high-cost capacity. Now again, this is an assumption with some uncertainty, but referring back again to our recent visit to China, the entire conference, also company meetings, our takeaway is that there is an increased awareness among Chinese players of the reduced demand growth also in China, and how that needs to be matched by lower rate of increase also of capacity.
Adding another layer of detail, let's look at the metal balances in key regions outside China, and we do see market deficits increasing in key consuming regions. Now most importantly, both the U.S. and Europe have an increasing primary supply deficit. After the smelter shutdowns in the U.S. last year, and before that, as well as we've seen in Europe over some years, this primary demand, this primary metal deficit is growing and needs to be met by increasing imports. The resulting trade pattern is that the U.S. is being supplied increasingly from the Middle East and from Russia, in addition to the traditional metal imports from Canada, while Western Europe is being supplied by Eastern Europe, but also increasingly from Middle East and Russia.
We're also seeing that this deficit is having some effect on the metal inventories. Let's look at the inventories in a little more detail as well. We see China's primary stocks having been largely stable in 2016, while total stocks outside China are decreasing. Now, this follows somewhat logically from my previous slides. The market deficit outside China is being met partially by metal exiting inventories. On the right-hand side, we see that total inventories outside China have decreased over the last years. We've seen a decrease of around 1.3 billion tons in the period from end 2013 through so far in 2016. Now we see that the decrease in LME or reported inventories is particularly clear.
This is caused, among others, by changes in the LME rules, affecting both the incentives to store in LME inventories, it's the pricing of storage, and its simplified procedures also to exit volumes. We see a decrease in inventory days, although they remain at quite high level compared to historic average, and the historic average is around 55-60 days. Now as I said, metal exiting inventories is going partially to consumption. As you see from the graph, as well, we can also see that a fair share is being moved into non-LME warehouses benefiting from lower rent and ease of exit. If you look at the left-hand side again, we see that the Chinese total inventories are showing a fairly flat development. We also see that the SHFE-registered warehouses have a certain somewhat decrease.
Now naturally, there is a lot of uncertainty regarding unreported inventories, so also in China. But we believe that there are certain limitations in that the Chinese metal oversupply cannot transfer one-to-one into exports. A large share of Chinese metal is being supplied to downstream processors as hot metal or liquids. Now this metal does not have an immediate export route if domestic demand is decreasing, and it would need significant casthouse capacity to be moved out as exports. Let's move over more to the cost side. Now the energy source for primary production varies between regions. The energy source into power generation is the most important differentiator for smelters. Now due to its power intensity, aluminum smelters have historically been located in regions where there is an access to abundant energy and relatively competitive energy.
We've seen it historically in hydropower-rich regions, so Brazil, Canada, Russia, Iceland, Norway, even parts of the US. We've seen also this happening in close to gas sources and gas-rich regions in the Middle East, and we've seen it more recently happening close to coal abundance in India and in China. Now different from the other main input factors for aluminum, power is a regional product, and it is regionally priced. Alumina, carbon, are traded globally, have fairly similar input prices for all players. So even though power prices are subject to certain globalizing effects through the fuel price, through the coal price, through the gas price, through LNG, it also remains with regional differences. As you will see on my next slide, these effects are currently at play, increasing the smelter costs in China.
Chinese smelters buying coal from the market, as well as smelters being on grid, buying power from the grid, are impacted by rising coal prices. If you look first at the pie chart on the right-hand side, we see that the majority of Chinese smelters are coal-based, and the power sourcing models can be divided largely into three categories. The smelters with captive power plants and captive coal. These are typically the lower cost Chinese smelters. These smelters are largely insulated from coal price volatility that we've seen recently. Their internal coal cost may differ from the market, and the market may not be an outlet for their coal. They may not have access to it.
Secondly, we have smelters with captive power plants, but they purchase coal on the market. These smelters are expected to be the ones most immediately affected by the coal price increases that we've seen over the last year. As a reminder, a coal price increase of about $1 per ton of coal translates typically into an increased aluminum cost of $5-$7 per ton. Thirdly, we have the smelters that purchase power from the grid. Traditionally, these are the high cost smelters, the fourth quartile ones. These smelters seem to be more exposed on annual contracts, annual price adjustments on their power, and they may not yet have seen the full effect or seen any effect of the coal price increase.
It's a bit up to the resilience of the current coal prices to what extent the coal effect will actually translate and in which order of magnitude the coal price will travel into their the power prices in the fourth quarter. On the left-hand side, we see how the coal prices have increased over the last year. Again, this is driven in particular by China. There's been a politically intended restructuring going on of the coal sector in China in order to improve the competitiveness of the coal sector. This has been, among others, by taking out or wanting to take out over capacity. We've seen restrictions imposed on the number of working days for mining companies.
Now, this actually affects across the board, also those that have captive coal are expected to be subject to this. We're also seeing restrictions in road and rail transportation. Some of this is temporary and seasonal, but some of it is legislation and more permanent. Now, again, going back to our recent visit to China, our observation is that coal price increases are expected to be tolerated just in some order of magnitude in order to restore profitability to the coal sector. At the same time, a large uncontrolled upward shift into the power price is probably unlikely, especially into the general power tariffs for general consumption. How does this affect the cost curve? Well, the global cost curve is shifting upwards by higher coal and higher alumina prices.
Now last year, we talked also about the shifts in the global cost curve and then looking at the downward shift that we've seen over several years. We saw that downward shift due to lower energy prices, lower alumina prices, lower premiums, and a strong dollar. Now this year, we're seeing an upward shift, and it's driven by many of the same elements. Alumina prices are up some $100 since the start of this year. Globally traded coal is up some $40 per ton of coal or $20-$250 per ton of aluminum if translated dollar by dollar into the through to the power cost. We've seen other costs also rising somewhat, but to a much lesser degree.
At the same time, the dollar remains strong and in particularly strong vis-à-vis the other commodity currencies. All in all, we expect somewhat upward shift in smelter costs and especially to be seen in coal-exposed regions. Going into and looking at prices and premiums. We've seen the regional standard ingot premiums having fallen back into historical levels following the peaks. We also see the all-in price for aluminum supported by currency effect if you look we look at it in NOK. The LME development has been stable to upward moving throughout the year. Starting from a low point below $1,500 per ton, we're now seeing prices upward or around $1,700 per ton, I think closing around $1,730 yesterday.
The upwards movement is a consequence of a market balance effect, so the capacity curtailments that we saw late last year, beginning of this year, but it's also better than expected demand in several markets. As was discussed in the CFO presentation, the dollar has historically proven to develop inversely with major commodities like oil, but also aluminum. The NOK, however, seems to be positively correlated with oil. Therefore, if you follow the red graph here or the red curve in the graph here, we see that even though we've seen LME being lower than what we have seen historically in dollar terms, translated into NOK, the currency effect actually moderates or even neutralizes the decreasing effect.
On the right-hand side, we see that since the 2014, 2015 peaks in premiums, the standard ingot premium has moved back to what we consider to be a historical level of around $140 per ton for European duty paid material. We see product premiums showing a similar development with extrusion ingot premium over standard ingot of some $200 per ton. Recycling is becoming more important as the generation of post-consumed scrap material is coming back into the supply stream. Aluminum consumption has been seeing steady increases over many years. Aluminum is used in a variety of sectors. Some uses with a short lifecycle such as packaging, some uses with medium or long-term or long lifecycles, typically automotive and building and construction.
Due to higher aluminum consumption over several years, volumes of post-consumed scrap are increasingly coming back into the metal flow and being returned, as you see here on the left-hand side, and the sharp increase expected also going forward. As such, an increasing share of the metal supply will be met by recycling going forward, and this is back to your question. However, we also see on the right-hand side that both recycled material, post-consumed scrap, and primary increases in primary are expected to be needed going forward. Let's move on to bauxite and alumina. Oh, sorry. Even more than for aluminum, the bauxite and alumina market really centers around China. We expect the Chinese primary production to be dependent increasingly on imported resources.
I like to use this most somewhat stylistic graph as an introduction. We see that in the period 2013 through 2016, the average annual Chinese aluminum production is some 29 million tons. To produce this, some 56 million tons of alumina is required. Over the last four years, we've seen that the Chinese domestic refineries have been able to supply some 93% of this alumina needed for primary smelters in China. The remainder needs to be imported. We see the 7% imported here in the mid-graph. Now, to feed the domestic refineries, China will consume around 140 million tons of bauxite this year. Around 90 million tons, around 64% of the total, is produced in China. Around 50 million tons need to be imported.
Looking forward, imports are expected to grow, and we also see that additional and new refinery capacity is being added in coastal regions. I'll come back to that in a later slide as well. Chinese domestic bauxite production is expected to level off, triggering the need for more imports. The Chinese bauxite deficit remains the main driver in the global bauxite and alumina market. Even though the major part of demand is covered by domestic bauxite, as I just mentioned, the volumes available are not sufficient to meet the demand. We see that we see the domestic in the green and light blue bars here. Chinese bauxite demand is foreseen to continue to increase in the coming years from some 65 million tons bauxite equivalent in 2016 to around 85 million tons in 2020.
With continuing depletion in own resources, the need for bauxite import increases. We see that in the purple bars here. This is happening at different speed for different regions. We also see some regions actually increasing their bauxite resources. We also see that the quality of bauxite is deteriorating, primarily in the north. This is adding additional costs to the refinery phase, so to the alumina. We see some regions set to run out of bauxite in 10- to 15-year period, and this raises the questions for the refineries in those areas. Converting to imported bauxite or moving the refinery to the coast or even moving the refinery outside China, all of this will add cost in the form of CapEx as well as logistics. Let's look a little bit more into the bauxite sourcing for China.
It's a fairly dynamic landscape. To supply its import requirements, China has gradually developed new resources and new areas. Australia traditionally, India, as well as Indonesia. The drive for new sources is particularly acute after the mineral export ban in Indonesia at the start of 2014. We do not see any signs that this ban will be lifted. On the left-hand side, we do see that Chinese players have been successful in finding new resources and continuously defying expectations in the market. After the Indonesian ban, we saw Malaysia coming on very strongly in the market, 2015 into 2016. In Malaysia, we also seen a mining moratorium being imposed in the early part of 2016 to enforce regulations and improve conditions around the extraction and exporting of bauxite.
The moratorium has since been extended and is currently valid to the end of this year. We also see bauxite being imported from a variety of other sources. Of course, we need to note Brazil one of those. In 2016, the most important new development is the emergence of Guinea as a major new supply source. We see here as the purplish bubble in 2016. Now Guinea has vast resources of bauxite. Western majors have been present in Guinea over many years, more than 40 years. In 2015 and into 2016, we've seen the SMB-WAP alliance establishing themselves with the mining operations and starting to export bauxite. Now this is the SMB and WAP is a combination. This is Société Minière de Boké and the Winning Alliance Port with strong Chinese ownership interests.
The estimated export volumes for the alliance in 2016 is in the area of 12-15 million tons, and the group itself has given a target for 2017 of exports and capacity in the area of 30 million tons. Now the setup represents low capital investment. This mining and barging concept is quite different from the classic large mine projects that we have seen elsewhere. On the right-hand side, we see the bauxite price development reflecting important market developments. Now, what I wanted to point out is that with the appearance of Guinea, the price effects of the Malaysian mining export mining moratorium seem to be quite limited.
To complete the picture, we also need to take into account that longer traveling distances means higher freight exposure, which we will look at in a little more detail here. Chinese bauxite imports are increasingly exposed to freights. The development is quite clear, and we see in the pie charts here from 2013 to 2016. The share of bauxite sourcing to China coming from Atlantic sources has increased from some 4%-9% in the 2013-2015 period to 24% year-to-date end of October 2016. This means that Atlantic bauxite is replacing Indian and Malaysian volumes. Indonesian, sorry, and Malaysian volumes. Australia remains the base load supplier, but Guinea is being established as the emerging new large supplier. Atlantic sources means longer freight routes, longer traveling distances, higher logistics costs.
We also see on the Baltic Dry Index here that freight costs also tend to have a certain volatility and of course, moving with an oil price development. I probably need to move on very quickly. How does this work into alumina? Well, high-cost Chinese refineries are impacted by imported bauxite prices. We expect new refineries in China of around 9 million tons over the next 2-4 years. Some refinery capacities being added inland, Shaanxi, Yunnan, Guizhou, taking advantage of domestic bauxite resources. But more importantly, most new build refineries are being added on the coast in Shandong and Hebei. These refineries are supplied by imported bauxite and are facing higher bauxite costs.
Looking at the alumina cost curve, the fourth quartile is dominated by Chinese refineries, as we see represented here. Now, the by far largest region is Shandong, around 20 million tons of refinery capacity. Compared with 120 million tons of global capacity, it's also one-third of the Chinese capacity of some 60 million tons. This is the region with the highest operating cost, again, caused by bauxite prices. It's also the region where most new capacity is being added. Again, this illustrates the fact that looking further ahead, domestic bauxite resources in China are not sufficient. China also continues to import alumina, somewhat lower levels than what we've seen historically, and reflecting the somewhat improved balance outside of China. Looking outside China, the alumina curtailments that we've seen over the last year are reducing the oversupply.
The alumina market has seen major changes over the last years. At this point last year, alumina prices were tumbling due to smelter closures. Few refineries were cash positive. Due to low prices, and the price actually bottomed out below $200 per ton in January this year, we saw alumina refineries being curtailed, both in the Atlantic and in the Pacific. Now, most of the Chinese refineries have since been restarted, but the Atlantic capacity of around 6 million tons remains closed. We've seen some additions in 2016, typically Ma'aden and Ketapang in Indonesia, again with Chinese ownership, having come on stream in 2016. However, overall, the development in 2015 and 2016 reduces the Atlantic long position and eases the reliance for the Atlantic region on imports to China, exports of alumina to China.
To sum up bauxite and alumina, we see alumina prices rising due to smelter restarts and cost inflation. The alumina balance has improved towards the later part of 2016, as you see smelter restarts and new capacity coming on stream in China. Prices are increasing, supported by strong demand as well as the cost push that we've seen both in bauxite and as well as in energy, as we discussed for coal for smelters. We also see an additional tightening of Chinese prices due to logistics effect, both temporary as well as legislative and possibly more permanent. We've seen production and sourcing hiccups as well due to environmental inspections that are ongoing in China. But it remains to be seen if we will see any permanent effect, either closures or higher costs due to environmental grounds.
Now, the alumina price, the PAX here in blue, reflects the market development that I have described. Oversupply. Need for tailwinds due to smelter closures end of last year. Rebound in aluminum demand into 2016 being stronger than expected and incentivizing restarts through the price effect. To sum up, I'd like to share just a long-term outlook and key developments. We see strong growth drivers across all segments, providing a solid demand outlook for the long term. The long-term demand for aluminum is expected to remain strong. It's based on fundamental drivers, the macro and GDP, and it's across all segments. In addition, aluminum will continue to profit from substitution and taking market share from other materials.
The substitution effect is expected to be strongest in automotive, but we also see it in the electrical segments. We see an estimated CAGR for automotive or for transportation of around 4%-5% for the period 2016-2025, and 3%-4% in the electrical segment. Main growth drivers being urbanization, substitution from copper in power transmission, more aluminum into transportation. We see somewhat lower growth in the building and construction segment, driven primarily by a slowing down in China. We expect a long-term growth of semis demand of 3%-4% for this period. As Rickard mentioned, this is somewhat down from our expectation last year, driven by a somewhat lower outlook on GDP and lower Chinese building and construction.
We see the growth in global semis, creating opportunities both for primary as well as for recycled material. We see the growth in recycling capacity, and the logical consequence is that more of the metal supply will be based on post-consumed scrap coming back into the metal flow. The forecasted CAGR of semis of 3%-4% allows for both growth in primary as well as in recycling. We expect recycling to grow in the area of 4%-5% on average for this period, and primary to grow in the area of 2%-3%. Now, if you look to the illustration here on the right-hand side, the semis demand growth to 2025 is expected to happen increasingly outside China, with India and other Asia being the main most important drivers.
To summarize my presentation, we continue to see macro and substitution effect supporting continued aluminum growth. We expect limited primary metal supply growth outside China and India. The global primary market is largely balanced, is expected to be balanced this year as well as next year. We do see a cost pressure due to especially coal price and power price increases. Recycling growth accelerating will increase or as with increased generation of post-consumed scrap. We see the Chinese bauxite import requirement continuing, and we see solid long-term demand outlook supported by strong growth across most segments. I would like to introduce Head of Rolled Products, Kjetil Eide.
Good morning. A few things have happened also in Rolled Products during the last year since our last Capital Markets Day. Starting with the commercial things, we've picked out some highlights here for you. We are very proud to have signed a long-term contract with Jaguar Land Rover early this year. This has been a milestone for us because we traditionally we have been very focused and concentrated on the German OEMs and JLR being intensively using aluminum to come in for Hydro as their second supplier and a growth supplier for them is important for us.
Also on the commercial side, we have entered into a long-term sourcing contract, the second and final part of that for our integrated smelter in Rolled Products, Rheinwerk located in Neuss, completing then the long-term sourcing until 2025 at very favorable terms. Beyond that, we have received a lot of recognition and awards from our customers from the market during 2016, which I will refer to. When it comes to capacity development, a couple of highlights. We completed the expansion of our hot rolling capacity in Alunorf in the beginning of the year. We had the official opening of the UBC center in Neuss in May this year.
A couple of months ago, we had the first production, the first coil produced at our AL3 line in Grevenbroich in Germany on time. Technology-wise, we last week signed a technology agreement with a company called Austin AI. It's a US-based technology provider within the sorting of materials and metals. We entered into an agreement, which I will revert to later today, on sorting of our automotive scrap, which is an extremely important step for us to become the leading supplier of recycling solutions to the automotive industry. Beyond that, we also have received other recognitions, like recognizing our body-in-white in aluminum for electric vehicles, which we were awarded the gold medal at the eMove360° fair earlier this year.
Finally, we also marked during this year one year of our internal Renew Rolled Products program, which for us has been a very important program the past year. We traditionally, when you talk about increasing volumes and output and performance, we traditionally talk about capacity expansion, investing in new capacity, investing in new technology. But we can achieve a lot by simply improving our own internal organization. Getting more out of what we have in the existing assets through working better with the organization and with the competence we have. That's why we launched last year a program called Renew Rolled Products, which is aiming at exactly that. It's aiming at lifting the cooperation across rolled products and increasing the engagement in the organization.
We do this with a focus on safety improvement, with a focus on improving our business system, and also simplifying processes. This is very much about people development and about leadership development. Renew Rolled Products is really for us, it's an enabler for us to deliver on our Better, Bigger, Greener targets. For us, better means improving internally our in-house processes, being able to produce at much better control of our in our operations with high degree of safety and also improving our performance towards the customers. We would like to use that performance and the position we are established with the customers in the markets to further improve our market position, increase growth in existing markets, and also go into new markets to have that as a growth lever.
On top of that, we also have strong growth ambitions in the area of recycling as part of our better, bigger ambitions. Greener for us, it really means to deliver to our customers in the market sustainable solutions and to be in the front when it comes to that, and also to be a key contributor to Hydro's ambition of becoming carbon neutral by 2020. Some achievements during 2016 when it comes to safety, which is the foundation of everything we do, we have had a significant improvement. Our employees' injuries have been reduced by 23% as of third quarter of this year. We have continued the improvement during fourth quarter, so we are on a good way to achieve our target for the full year of 4.0.
Looking at the contractors, we have even improved more than that. We have reduced injury rates by more than 50% and the combined performance on TRI, what we call TRI rate, the second half of this year, we are at a level below 2, which is our long-term target. We are really proud of that, and we have a good development. When it comes to shipments, first of all, as Cathrine has pointed out, the real growth segment for us, automotive, we have increased our sales of body in white to the automotive industry by 30% this year. Not through increasing capacity, because we haven't had new capacity this year. It's simply by improving the utilization of the existing capacity. Before we really start counting volumes out of the AL3 investment.
In total, shipments out of rolled products this year increased by 5%, which is beyond the general market increase. We have all in all taken a good piece of additional market shares this year. EBIT financial results, I guess some of you are a little bit interested in that also, has been lifted during 2016. Counting the four last quarters compared to the previous last four quarters, we have an improvement of 28% on our underlying EBIT from the rolling operations. Eivind explained to you how also the impact from our integrated smelter with the LME impact or the full price on the primary metal impact has had also on our EBIT for 2016. This is just reflecting the rolling operations.
Finally, as I would say, a confirmation of achievements we have done through the Renew Rolled Products program, we have lifted the engagement in the organization by a significant 8 percentage points in the last employee survey, bringing it up to 83%, which is in line with the best 10% of companies who are measuring employee engagement in this way. I think we really have a belief that we have done a good job in terms of improving our organization to support our Better, Bigger, Greener ambitions going forward. When it comes to the different market segments we are in, I think you know them. I'm not going to go through that in detail. We, of course, want to grow the segments which are most attractive growth opportunities.
Automotive is obviously a growth area for us with also new investment. We are also growing selected other parts of our portfolio. Within each market segment and product segment here, we will continue high-grading. We will take out products where margin opportunities are less, and we will increase within niche or parts of each end product segments where there are better margin opportunities. Altogether, of course, and lift the total performance by optimizing across the entire product portfolio. That is a key part of what we are working on in Rolled Products. As an illustration of that, looking then, we don't reveal all the detailed volume numbers to the market, but we can show that improvement in automotive output.
Our target for 2020 is 24% of our sales should be towards automotive by 2020. That is a doubling of its the relative share of automotive products compared back to 2010, when we had 12%. Those of you who were here last year would also maybe remember that we then presented an ambition of 20%. We have stepped up our ambition. We see further growth opportunities within automotive, within attractive margin segments. We have gotten confirmation also in the contracting for our L3 investment that we are having an attractive product offering to the market. Also within the other product segments, we do high-grading, like for instance, the special products area, which altogether is reducing its relative share of our total product portfolio.
Within that, we are increasing what we define as strategic special products, where there are niche segments with good margins, which requires exactly our competence in quality and advancement and service levels and so on. Within each of these segments, there is a further high-grading. For us to be able to deliver on our better, bigger, greener ambitions, one key factor is ability to differentiate in the market. Part of doing that is improving our quality, it's improving our service level and service concepts, and innovation speed to the market. There are different ways of measuring how well we succeed with that. We measure it, of course, with volume development and margin development, which is a little bit of a lagging indicator.
We're also asking our customers through frequent surveys and feedback from customers on our performance is showing quite promising development. Further on, we also achieved a lot of recognitions and awards from customers, from industry associations and other industrial bodies. We've listed some of them that we achieved this year here. For instance, Ball, which is the biggest can maker, they awarded us as the best supplier of body stock material in Europe last year with the organization which is now called Ardagh, which was spun off from the Ball and Rexam merger process. Another example, at the eMove360° trade fair in Munich earlier this autumn, Hydro was awarded for our development of all-aluminum body concept for electric vehicles, which we developed together with Sapa and RWTH Aachen University.
A concept which has brought the weight of a full body down to below 200 kilos. We got a gold medal in the competition with 10 finalists, and we're really happy for such a recognition. It underpins our strategy to be the leading supplier of aluminum to the automotive industry in Europe. A couple of examples when it comes to service and quality. These are examples from our litho business. Litho plates are under extensive pressure and stress during the printing process. With the heavy alternating bending during the printing, and particularly the large plates, which is a growing part of the litho business, have a tendency of cracking.
We introduced a new breakthrough alloy, which has shown a significant reduction in the cracking process during the printing process, and up to 90% reductions of cracking, which of course has significant improvements in productivity for our customers and also the quality level. We have achieved this through a deep close relationship, not only with our immediate customers, but also their customers, to understand their process and how we can improve our product to support their success. This, we expect, will be the new standard for litho sheets in the future and supporting our number one position globally in this very demanding product.
Service concept, we have reduced our lead time to deliver litho plates to the U.S. market by almost 50%, since 1995 until today, from 11 weeks down to six weeks, significantly reducing the disadvantage of supplying litho plates across the Atlantic, and further increasing our competitiveness and solidifying our leading position on a global basis with this very demanding and good margin product. Couple of examples of innovations from the automotive side. We introduced this year a copper-free material for the header tubes in heat exchangers. These header tubes are typically produced containing copper. When they are connected with the multi-port extruded parts in a heat exchanger unit, these two parts are not galvanically compatible.
What we have been able to do is to develop a solution which is completely copper-free, which has then reduced the corrosion problem in this interface, which again has reduced, of course, the quality issues and the risk for field failures of these units as they are installed and used in cars. That has been a significant step change and a breakthrough for us within the heat exchanger business. I mentioned last week we entered into an agreement together with a technology company in Austin, in the US. It's called Austin AI, this company. They have developed first phase of a laser-based sorting technology to be able to sort 5xxx and 6xxx alloy material, which is what you use then to produce automotive body parts.
This is very difficult to actually be able to sort that, technically, and automatically. This technology, we believe, is going to be a breakthrough in that, which will enable us to go to our customers with a solution which is far beyond what they have today, and what others can offer. We have made a decision to make a pilot scale facility to establish that in our R&D center in Bonn, where we will further develop this technology, and prove it. Once proven and successful, we will then take the next step towards our customers. Very interesting opportunity for us. Further on recycling, the UBC center in Neuss, it was opened officially in May this year.
We are in the process now where we have a ramp-up, but we are operating on a stable level, but we are behind our target when it comes to speed of the ramp-up. This is very much due to technical issues in certain parts of this line. The UBC center is a fully integrated line, starting with the shredding of the material, mainly then used beverage cans, but mixed with all other type of potential material. We are shredding it, and then the fractions out of the shredder is then automatically sorted. We get out the aluminum fractions, which are again then automatically brought into a delacquering unit, and from the delacquering unit directly into a melting unit.
All this complete integrated chain with the most advanced and new technology is working fine, but not flawlessly. We need to find some areas of improvements. We need to make some modifications to debottleneck to bring it up to the ramp-up speed that we originally targeted. We are in the process to do that. This delay is what is causing us also to be a little bit delayed on our Rolled Products achievements for 2016. Another big project that we just started is the automotive line in Grevenbroich, and I think we will give you an opportunity to have a look at what we have done in this project the last 20 months within one and a half minute. Please. Yeah.
This was the Automotive Line Three. It's started first week of October, exactly on time. It's on budget, and it has been built with a good safety level. We are now in the ramp-up process, and in the qualification process. This line will have to be qualified with all our major OEMs, starting with the European OEMs. Both line and every single product needs to be qualified. We are also testing the quality. This will be able to produce superior quality Body-in-White material to the automotive industry. As Svein Richard Brandtzæg earlier explained, there are certain elements of this new line which is superior to what is existing today in terms of the quality, in terms of the performance of the material on the test shop alliances.
So far, qualification and quality testing here has been progressing well. Finally, sales contracting has so far been very positive, and we are well ahead of original plan. Our Better Rolled Products program is progressing. We are on track when it comes to our 2019 goal, although we have some delays on the UBC in contributions from the UBC center in 2016. Operational performance improvements is ahead of plan. That concludes my presentation. Of our six midterm targets that we presented last year, three of them are unfortunately impacted by the delay in the UBC center. It is the Better Rolled Products, it is also the UBC center itself, which is a separate target, and our growth of post-consumer recycling.
As I said, we will catch up on that. We are certain about that. When it comes to the one step change per year, we have delivered one step change, which I just explained. Beyond that, we have in total four major step changes for us during 2016, and we have a lot more in the pipeline for the future. The body in white investment is on track, so that is green. When it comes to safety, as I said, we are trending now at a level with a TRI rate below two the last half year. I think that we will be in a good position to deliver on our long-term target of below two by 2020 altogether. Thank you.
By that, I would also like to introduce my colleague, Hilde Aasheim, responsible for our Primary Metal. Please go.
Thank you, Kjetil, and good morning to all of you. 2016 has been another interesting year for Primary Metal. Obviously, the decision to build the Karmøy technology pilot was an important milestone for us. Since then, we have been busy with constructing the pilot and now also getting ready for a startup in the second half of 2017. With the technology pilot, digitalization initiatives are also now gaining momentum, which is quite exciting because I believe it will change the way we work going forward. We have also been focusing on cost improvements and this year we will finalize the $180 program for the joint venture smelters, which came a couple of years after we introduced the $300 program for the fully-owned smelters.
We've also been active in the market, and we have intensified the work in terms of product development towards our advanced customer, in particular in the automotive segment. I was also very happy when I could sign a multi-year contract with our largest customer, Sapa, and confirming the good relationship between Primary Metal and Sapa. Together with Energy, we are working hard to secure competitive power contracts for the smelter portfolio. As has been said, that is the major differentiator in terms of being cost competitive when you are in the smelter business. This year we have signed a new contract, the one Tereho, for the Norwegian smelters, from 2021-2039. Then we have heard about the prospects of recycling.
We are now turning our remelters into recycling units, and we have finished a project in Clervaux, in Luxembourg this year, where we now will be able to dig deeper into the scrap pile, bringing in more post-consumer scrap, putting it on the furnace, and produce so-called recycle-friendly alloys. When it comes to the strategic priorities for Primary, it stays more or less the same as last year. The better part, it's about safety, it's about cost, and it's about superior products. When it comes to B&A, it's about to squeeze more out of existing assets, with the creep program of getting 200,000 tons out of the existing capacity, which is about 2.1 million tons of electrolysis capacity.
It's also obviously about successful build and ramp-up of the technology pilot, which would add another 75,000 tons. Bigger is also about growing now our recycling business with interesting prospect, both from a profitability but also from the greener aspect of bringing back the used aluminum back in the loop. Greener is also obviously about we're working with our own emissions and then also working with the customers to assist them in reducing their footprint. Some of you that have been to the capital markets day for several years have been hearing me standing here talking about cost improvements. This is really something that has been our focus since 2008. It has been the $300 program.
It has been the joint venture program. All in all, we have improved in terms of cost of NOK 3 billion in real terms. We have been working in areas where we can influence. It's not about raw material prices, it's not about the market. It's really about improving the operational excellence, as well as taking out waste in the whole chain, reducing fixed costs. If I include some few portfolio changes in that period of time, we have been able to reduce our fixed cost per ton by 50% over that period. We have also been able to improve our productivity measured in ton per employee by 30%.
The results come from a performance culture, with dedicated teams combining our total competence into a structured, systematic way of working in our aluminum business system, and is to turn competence into cash. I'm really proud of my organization, and the recent employee survey showed us that, we are really world-class when it comes to performance, but also on engagement index. This slide shows, some presentations has also pointed to that, the improvements, is visible in our numbers. To the left, we see the all-in implied primary costs, which is derived from all the all-in prices, less, the EBITDA per ton when I include also Qatalum.
To the right, you see the position, our position on the CRU curve for 2016, which shows that we are competitively positioned in the best part of the second quartile. In these charts, we also have influence from currency and also market effects. When I'm looking at the cost curve, my concern is that it's getting flatter and flatter. That is why we have to continue to improve in terms of sustaining, but hopefully also ideally improve the cost position. Because there is only slight increase, slight changes on the currency, or power cost, that can move us in the wrong direction.
That is why we celebrate successes, but we are going forward with more improvements, looking for areas where we can find new ways of working. A couple of years ago, we looked at our electrolysis capacity, and we believe that we can squeeze out another 200,000 tons of the existing assets. The first 100,000 tons within the next five years, we have known technologies. We have tested technologies that we could install on existing lines. We are well on the way to do that, and we have achieved 35,000 tons of that program as of 2016.
It's about improving the cathode, which is the pot, in order to be able to put more amperage into the cell. It's about increasing the anode in order to bring the current into the bath, into the electrolysis process. It's about combining our competence to be even better than we were yesterday. To do that, as I said, we have to have a performance culture. We have to use all the competence we have, not only with the technicians and the R&D people, but turning that competence into the hands of the operators.
When I was standing here last year, I indicated that to do this first 100,000 tons, we needed some CapEx simply to block out some of the infrastructure of the equipment. Then I indicated roughly $1,000 per ton of this first 100,000 tons. We believe now that that is on the high side, which is then quite cheap capacity if you compare that to new builds. It also the fact that we can increase capacity from existing assets reduce also the unit per ton cost of the capacity. That is going well. I also have to say that we are maturing these projects, and every project have to have a profitability on its own fundamentals.
It has to be a market for this volume, and it has to be a profitable investments once we are going towards this path. When it comes to the 100,000 tons by 2020, we're really waiting for the spin-offs from the pilot. The pilot consists of 25 different technology elements, and the good thing about the pilot is that we can take some of these technology elements and invest in existing assets. It's also about waiting for the new novel control platform, which will really help us to be able to squeeze more metal out of the pot. That will come once we have the technologies verified in the pilot.
The first years of the creep is built into the NOK 1 billion new target that we have set for ourselves up to 2019. We have delivered. We had a target of NOK 400 million. We're slightly below that simply or mainly because we had the unfortunate power outage in January this year at Årdal. We had to take down 10% of the capacity in order to save the plant during a very unfortunate power outage during a storm on the west coast of Norway. We are back on track now.
We are at full speed at Årdal, and we are at a state now that I can say that I feel confident that we can be able to deliver the next years up to accumulated NOK 600 million for at the end of 2017 and then on a good way to deliver the NOK 1 billion kroner at the end of 2019. Technology and improvement goes hand in hand because this is not primitive cost-cutting, what we're doing.
We're really using our competence to accelerate our performance. That is why we all wait now for the advancement of the pilot in order to be able to start to test the new technology that has been working in six cells for 5-6 years. As has been said previously, we reckon that we will be at 70% physical completion at year-end. The project is on budget and on time. The current people is now busy to prepare for the operation and to be able to take over at the second half of 2017. As I said, we're really excited about the pilot.
First of all, because we will demonstrate to the world that we are able to operate a much more productive cell with a much less energy consumption and with the lowest footprint in the world. We're also excited simply because we can take out effects from the pilot to existing assets. As Cathrine has explained, the industry conditions is so that to add new capacity these days is not on our agenda, but we want to squeeze out the most from existing. That is why we have now mobilized a program and an organization to now start to tailor make spin-off packages and solutions to the other electrolysis line in the system so that we are ready once we have the different technology elements verified.
I'm also very excited about the technology pilot in the sense that this is our first step on the way to a digital future, and we talk about Smelter 4.0. The new pilot comes with a new novel control platform. That is a prerequisite for the success of the pilot, simply because we are going to tame, and I use the word tame. We are going to tame the forces in this giant cell. We simply need to have a new control platform in order to be able to operate the cell with much higher precision and with much less manual work than in the past.
This will be a cell with much more sensors, which will give us feedback continuously. Much more predictive systems, which will be a decision basis for the operators to operate the cell. The predictive systems, the learning machines will tell us the progress of the cell, the prediction of the cell. Big data will actually help us to reduce manual work to actually take the anode effect when it happens and not waiting for the operator to come and tend the pot. All these new digital way of working will be built into to operate almost the cell autonomously.
This will also change the role of the operator and even the engineer in terms of getting more feedback, getting decision support, and to be able to operate this giant in a controlled way. When we have this new novel control platform installed, I foresee that this will also be able to be implemented at the other smelters. One time in the future, my vision is that we will be able to connect the plants, all the plants into one virtual organization. The new pilot will also come with more automation in terms of pot tending machines. It will come with more automated transport system. It will really be fantastic to show more about that once the pilot is up and running. It's not only about cost in Primary Metal.
We are in the market with almost 2.9 million tons when I include also the full volume from Qatalum, which we market and sell, and also volumes from Froco. We really have to work, and we're really working on positioning ourselves in the market, differentiating ourselves towards competitors and really getting values from our superior products that we produce. In terms of what we produce, we have a strong position in the so-called value-added product segments. We are producing only 18% standard ingot, and that means that most of what we produce is tailor-made to specific customers with a specific alloy. The share that we have of value-added products is 82% of our total production of casthouse products.
To the left, you see how we compare with our other peers in terms of value-added products. The value-added product strategy has created substantial values over the years. Here I pulled out the CRU Casthouse value creation curve for 2016, where you see we have a very strong position in terms of creating values, producing value-added products. The net realization margins that is illustrated here is derived from the product premium, and then you deduct the freight, the marketing cost, the financing cost, and the duties. This net realization margin is an extremely important part of the competitiveness of our smelter cost position because it's a credit to the total cost of producing one ton of metal.
It really supports the competitiveness of our cash cost position for the smelters. As I said, the value-added product strategy has created substantial value over the years, and every year we are revisiting our product portfolio in order to optimize the portfolio of products as well as maximizing the values that we are creating in the market. We work hard to sustain and develop that position going forward. In terms, in order to differentiate ourselves further compared to the competitors. The core differentiators are obviously to come with a superior product and have a world-class excellence when it comes to delivery performance.
We will simply not be able to sit with the most advanced customer to discuss with them how can we develop together if we were not able to deliver each time on time with the quality that the customer need. Because then we are in a position to bring our technical experts together with our R&D experts to discuss what next. What does the customer need? How can we provide them with new alloys for new applications? Right now, we work hard towards the growing automotive segment. We are working to produce sheet and ingot for the body-in-white. Right now, we're working on extrusions solutions for new bumper system for the car.
We are working, like, Sven Richard Brandtzæg mentioned, on crash alloys to create a more safe car and to take the energy out of the crash in the car. Very interesting development and good growth in the automotive segment. We also see that customers are coming to us and want low carbon footprint metal. With the superior position we have in Hydro with both world-class when it comes to direct, which is emission from the cell, and then indirect using hydro power for two-thirds of our production, we really have. We believe that we have something that can differentiate ourselves also going forward.
We also have customers that want recycled metal, and that is why we are quite excited now to take the 7 remelters that we have, 5 in Europe and 2 in the U.S., to turn them now into recycling units. That is also why we invested in new novel sorting and shredding technology last year to dig deeper into the scrap pile to bring the sorted and shredded scrap into the furnace. Here we have technology. We have cast house technology in Hydro, and we believe that here we can really create a unique position. On that topic, I would like to invite my head of recycling, Roland Schärpenberg, to talk a little bit more about the recycling journey.
Thank you, Hilde. Good morning, ladies and gentlemen. Let me tell you something about recycling. Let me tell you about how we close this loop and make this green as it is shown here. Let's look at the right-hand part of this slide first because it shows the value chain of aluminum, starting with bauxite, alumina, primary. It goes into cast house, then via semis into final products. In all production steps, process scrap is generated that today is already recycled, and we call this remelting. What goes into the final products ends up in the big pile on the right-hand side, which we call the metal bank. This metal bank is going to increase by the year 2020 to more than 1.1 billion tons.
This is where we are more and more harvesting from with post-consumer scrap recycling. The post-consumer scrap collection that is shown on the left-hand side of this slide, this will grow significantly going forward. We expect that it will double from today to 2030 to almost 30 million tons. Because of that, we have done a strategy review, first of all, also to look at the process scrap to close loop, especially with our customers, but also then to take a higher share of the post-consumer scrap market. This is then combined with our leading positions in extrusion ingot and in rolled products. To illustrate how the recycling process works, we are showing here how do we get from scrap to the finished product.
First of all, the scrap arrives in our plants. It's process scrap. It's preconditioned, post-consumer scrap. We do the quality check, the quality control in the receiving department. The material is then stored in the right place to be consumed at the right time. The charge planning calculator, it's a Hydro in-house developed program, calculates the optimum raw material mix, optimum for the product quality, but also optimum for the metal cost, which by the way, make up for 90% of the cost in a recycling plant. When we have done that, the material is charged through the melting furnace and then alloyed in the casting furnace.
The final quality control then tells us whether the metal is ready for casting and after casting, we do the homogenizing and cutting and finally the packing to the finished product that will be shipped to the customer after final inspection. All these processes are standardized in all our plants around the world. We work in all plants in the same way. We collect all data from all plants in our production system to be able to benchmark and identify where we are lagging and where we are leading to learn from each other. This shows here the seven recycling plants that Hilde mentioned, plus our scrap processing plant that we acquired last year. The seven recycling plants we consider as being benchmark in industry.
Because we have invested there in high automation, we are very efficient in melting and casting when it comes to energy consumption. With the high automation, we are also lowest in manning. The standard average plant out of these seven plants produces 80,000 tons with 50 people. By the way, 50% of the scrap that recycle in these plants is process scrap. This is customer scrap for us. This explains that, of course, this business we will continue to run, we will continue to provide the service, but it explains also why this limits us somehow to some extent on the post-consumer scrap recycling. Looking at the strategy, of course, it is key for us to build on our leading capabilities.
As I said before, we consider us being industry benchmark when it comes to safety, environment, quality, but especially productivity and cost. This benchmark position we need to maintain, and we will maintain this by fully introducing our AMBS, our Metal Business System, which is building on continuous improvement. When it comes to recycling, our main ambition, of course, is getting closer to the source of the metal, especially closer to the post-consumer scrap. This is why it is essential to either invest into scrap processing, because we need to upgrade the quality of the scrap to make it furnace ready, or to form alliances with our suppliers, which of course is the preferred option. In order to do that, we need to enable our suppliers, and this is why we are also investing into technology development.
By getting closer to the source, we are securing the raw materials for the future out of the growth that we have shown before. Combining this then with the increased development and sales of recycling friendly alloys and products, this will help us to meet our targets in the long term. It will lead to optimizing the metal margin, and finally, to a very good profitability of this segment. Looking at the key drivers that we have identified as part of our recycling strategy, it's on one side, the post-consumer scrap recycling, where we are planning to double to the year 2014 to 150,000 tons in 2020.
We are lagging a bit behind, but the reason for that is we took the conscious decision at the beginning of this year with dropping sales premiums to go shorter on the raw materials and also to free up cash with that. By that, we were able to maintain a good profitability in a market with falling product premiums. The other thing is, of course, and I mentioned this before, the recycling friendly alloy sales. There we have the ambition to lift the recycling friendly alloy sales to 350,000 tons by the year 2020, which is a 40% increase from when we started in 2014. With this, we are confident to deliver on our earnings targets.
On the right-hand side, you see that we were able to manage the last 5 years a 15% return on capital employed, on average. This is, by the way, almost double the results from the 5-year period prior. This concludes my part, and I would like to get back to Hilde Aasheim.
I just will sum up our achievements towards the midterm goals. Safety first. We always start with safety. The TRI below 2 is the target for 2020. Primary metal has been at 2 the last 2 years. This year we are a little bit higher, that is why we are green. I believe that we strive every day for an injury-free environment, and I hope to be able to achieve the target of below 2 at 2020. I already mentioned the improvement program, so we are on our way to the NOK 1 billion, slightly behind this year due to the Årdal. Also here, I'm confident that we have a good speed towards the NOK 1 billion. Technology, the creep capacity is well on the way.
The pilot is also in good progress. As was just mentioned, we have a target of 150,000 tons or to use 150,000 tons by 2020. Due to market effects, we are a little bit lower this year, but also here we are progressing. This is Primary Metal's contribution to the better, bigger, greener aspiration for Hydro. Thank you.
Thank you, Hilde. Can you please stay on stage? Roland and Kathrine, can you please rejoin us here? We have time for a very short Q&A before we go to the break. Again, microphones on the side, and please state your name and company if you have a question.
Hi again, it's Danielle Chigumira from UBS. One question on rolled products. You're now targeting the autos to contribute 24% in 2020 and from 20%, but that's in terms of volumes. In terms of profitability, could you give us any sense in terms of how much autos currently contributes and what that might look like in 2020? The second question is on recycling, and it kind of spans both the market side and the primary side. What is stopping from a global perspective a bigger increase in recycled volumes? 'Cause we're only got, I think it's 3%-4% total growth in recycling. Is it a lack of supply of post-consumer scrap? Is it a lack of capacity? Is it processing technologies? What's stopping that from accelerating faster?
Okay, maybe. Is it on? Can you hear me? Yeah, you can hear me, even if I don't think this is on, but we never comment on earnings per product margins or, you know, how much we earn for every any single product. You can be assured that when we grow in automotive, it's because it's more attractive also to grow there than to grow other places. Obviously, there is a volume growth, but it's also more attractive margins in that portfolio than the average of the rest of our portfolio. All else equal, you should expect then that the earnings through this relatively higher share of volumes to automotive will improve. When it comes to recycling, I think we can both comment upon that.
Obviously, it's a little bit of everything what you're saying. There is not a lack of scrap. There's not enough scrap available, but we need to make sure that you are adapting the scrap availability to your needs, technically speaking. That's why we work with recycling-friendly alloy development to better adapt our needs for final products to the availability of scrap. It's also a question of course, making money. We don't want to invest in capacity to take more scrap, if we don't see that as a viable way of improving return or getting a good return on those investments.
This is taken stepwise through technology development, by being more efficient in sorting, bringing down the cost, and better able to fit the scrap qualities to the needs of the final product.
I can add to that. We have a quite an aggressive plan to refurbish the plant in order to be able to take in post-consumer scrap. You have to understand what you're doing. You have to have the capability in each of the remelters. Once we have done that, and we see good profitability, then we have to look out. Right now, we have quite a busy agenda to take advantage of the capacity and the facilities we have.
Hi there, Henry Banner, Platts. Two questions on my end. First of all, regarding the U.S., well, the growing U.S. supply deficit over the next few years, what does Hydro see as the major prospects for supplying a market that will, you know, increasingly be in such a deficit? Do you think there's any value in mothballed assets domestically? Or are there any other structures that you would see to try and supply this market going forward? The second is on Chinese bauxite imports, and just what your thoughts are on the increasing amount of exports from Brazil to China over the past year.
I think I saw that from Trombetas through to China already by June, the amount of exports had already risen to higher than 2015 levels. Is that something you're particularly looking at, that supply route? If not, what do you think is driving it other than closing smelter capacity in Brazil?
If I can start by the latter, because you're kind of answered your own question, I think. The Brazil route, to a large extent, is a reduced need from other players in Brazil, and therefore an export of bauxite from Brazil to China. I think that would be the major part of that stream. Regarding the U.S. metal deficit, as I said, it's currently being supplied increasingly from Middle East, Russia primary. We also see some increase, of course, of semis.
If you look at the mothballed assets in the U.S., that, of course, would be a combination of prices and costs for the U.S. market, as well as to the extent that these assets are actually able to be brought back into production, which has the tendency to vary with or reduce with the amount of time they've actually been out of production. I won't speculate in what other players might do with assets that have been curtailed. I do believe it's a fair assumption that we will see continued high exports to the U.S. from the same markets as I mentioned, potentially also some from Europe.
Is Russia in Europe?
Sorry?
By Europe, sorry. Do you mean Russia?
Yeah, Russia is.
Yeah.
is a large exporter to the US already. There's an island between mainland Europe and the US. You might also see some semis going that route, potentially. We have seen some increases also in semis from China, as I mentioned, with flat rolled products on can stock side as well.
Okay. Thank you. That, I think that concludes the Q&A for now. Now we will have the second break of today, and I kindly ask you to be back here at 12:10 P.M.
Ladies and gentlemen, it's time for the third and last session of today. I would like to introduce Executive Vice President and Acting Head of Bauxite & Alumina, Eivind Kallevik, to the stage. He will also be joined by Simon Storesund, which is Head of Commercial in Bauxite & Alumina.
Thank you, Stian Hasle. Today I will go through the main developments that we've seen in bauxite and alumina since the last Capital Markets Day. The main focus will be on the very strong operational improvements that the BNA organization did realize in 2016. As Stian Hasle said, I will be joined by our Head of Commercial, Simon Storesund, who will give a little bit more of a deep dive into the commercial efforts as well as priorities that they do in order to help BNA realize the true value of the world-class assets that we have in Brazil. Now, if we start with the key highlights since our last Capital Markets Day.
2016 has really been a year of accelerating performance for bauxite and alumina, with overall solid improving operations as well as good progress made on several other targets that I will come back to. Alunorte saw an all-time low implied alumina cost of $183 per ton in first quarter this year. In addition, we've seen record alumina production over a period of 12 months, and then finally reaching the nameplate capacity of 6.3 million tons, and then in the third quarter, producing 6.5 million tons on an annualized basis, well north of the nameplate capacity. The largest sustaining project that we have, the new red mud disposal area in Alunorte, is also progressing according to plan, as the new press filters were also commissioned in the third quarter this year.
Paragominas continued to deliver high bauxite production, well above the nameplate capacity of 9.9 million tons, completely in every quarter during the year. We also completed two major maintenance projects in Paragominas, replacement of the gear in the ball mill, as well as replacing a certain part of the pipeline. As you know, during the year, Hydro and Vale did negotiate the acquisition of 40% of Vale's 40% share of the Alumar mine. This transaction did not come through. From a physical side, we're in the same place. We still have the 40% offtake from Alumar built on the evergreen contract we have with Vale. Also, on the CSR and community perspective, we are very much aware of the important place that we have in the society in Pará and in Barcarena as such.
One of the examples is that we do participate in the urban development plan in Barcarena, the location of the Alunorte refinery, as well as the Albrás smelter, where we support the developments of the city and the infrastructure. All in all, it's been a quite interesting year for Bauxite and Alumina, with several strong accomplishments. Now, the deliveries in B&A in 2016 takes us several steps closer to realizing the long-term aspiration of better, bigger, and greener, which is Hydro's response to strengthen our competitiveness and ensure value creation in a fast-developing world. Safety is, of course, always on top of the agenda. What is more important, and equally important, is that safety results is typically also a good indicator that you have good operational control that we've had in 2016.
We will continue to drive the ambitious improvement efforts to ensure that we maintain our world-class operations. As Simo will come back to, we will continue to improve the commercial performance even further, for example, by pricing both alumina as well as bauxite on its own fundamentals. On the bigger side, we will continue to secure and develop our raw material resources, again, for future decades to ensure Hydro's self-sufficiency when it comes to bauxite. We have some of the best bauxite resources in place in Brazil, as well as one of the best refinery projects in the CAP project. As we speak, we are also working to debottleneck Alunorte to lift the production capacity at Alunorte up to 6.6 million tons.
HSE and compliance is of course an extremely important point for Hydro, and it is very, very important for us when we operate in Brazil. This is clearly the focus to operate for us, and one of the key issues that we focus on is our ability to reforest mined-out areas. Through good improvements in 2016, we are very close to reaching the 1-to-1 target that we set for 2017, and we're also working on closing the gap on the previous lack of reforestation. If you look to the improvement programs, after we finalized the From B to A improvement program last year of NOK 1 billion, we did announce a new NOK 1 billion target ambition at the last Capital Markets Day.
We've had a very strong development so far in 2016, and we're well ahead of the annual target of NOK 500 million set for this year. Our estimate for 2016 is now NOK 650 million, backed by the very strong operational and commercial efforts carried out in the organization. We also expect the fairly strong performance to continue into next year, and we plan to deliver close to NOK 1 billion already in 2017. This, of course, doesn't mean that we're done by 2017, and will not aim at improving in 2018 and 2019. We will, of course, continue to look into improvement measures and what we can do beyond NOK 1 billion, but this we will communicate to you at a later stage. For now, we maintain our target of NOK 1 billion through 2019.
There are several elements which contributed to our realization of the better ambition. One is lifting the production levels at both new plants above nameplate capacity. That contributes roughly 50% to the program. Optimization of energy mix, improving raw material efficiency at Alunorte accounts for about 25% of the target. The remaining 25% comes from a strong cost discipline to reduce fixed cost, which is, in particular important in Brazil, given that it is a high inflationary country. Also, commercial efforts aimed at lifting margins, as well as selling bauxite out of Brazil at attractive prices and logistical optimization are important factors in this term.
If we then go to Paragominas, we have now for a long period of time been able to continue to produce at high levels well above the nameplate capacity of 9.9 million tons. The production increase of some 5% that we've seen so far in 2016 compared to the already strong levels that we saw in 2015 has been achieved primarily due to the implementation of the Bauxite & Alumina Business System. Which is very much a product of knowledge sharing and help from Primary Metal business area. Proves that this is one of the important factors and benefits of being part of a truly integrated global company. This is very much about the way we run the daily operations, and it's really about establishing good control on all the assets and all the individual production processes at the plant.
The main elements have really been to contribute to the stable and good production levels, have been improved equipment conditions, as well as revised and newly implemented operating standards. That improves the process control and improves the productivity. During the year, we have also been able to improve the ore quality control, meaning that we are much more able to be precise when to stop and when to start the mining, also on how deep to go in the different levels. That improves the ore quality, reduces the waste, and reduces operating costs. The large sustaining project we have in Paragominas, the building of a new tailing dam, is well on track and on budget. We spent roughly 50% of the CapEx so far, and we'll spend the remainder part in 2017.
With the existing tailing dam, we have finished the last heightening of the dam. The construction has also been done on track and on budget. For the new tailing dam, we do expect that to be completed in the second half of 2017. As I mentioned, responsible and sustainable operations are the way we do things in Hydro. Reforestation of the used mining areas is therefore of very high priority for us. In 2016, we have made much better progress than planned in terms of reforesting large areas, meaning that we are very much on track to meeting the one-to-one 2017 target. We are also well on track to meet the closing of the reforestation gap by 2020. If we turn to Alunorte.
I'm very proud to see the quite tremendous achievements that we've seen in this refinery, both in terms of production, but also in terms of operating costs. Last quarter, we did report a record production at Alunorte at 6.5 million tons of an annualized speed. Also very happy to see that so far in fourth quarter, we continue to produce somewhat above nameplate capacity. There's been a steady improvement trend in Alunorte now for several quarters, following several years of challenging operations. This again is the result of very systematic efforts over several years through our standardized production system, Bauxite & Alumina Business System, continuously improving the processes and using the complete competence of the Alunorte organization to take out waste in the production chain and to set and reach new goals.
The first has really been to stabilize the production and reduce the daily variability by controlling every step of the value chain. This has been necessary before we can now start to look into lifting production to the next level. The next steps, as I said, will now be to continue to debottleneck the plant and lift it to 6.6 million tons in the next couple of years. At the same time, we've seen a good improvement in the cash cost at Alunorte with a new energy mix and an overall lower raw material consumption. The rebuild of the coal-fired boilers is now complete, and we see a significantly lower fuel oil consumption in 2016 in favor of coal. This is improving the energy costs with approximately 8% compared to a year ago.
The next step of improvement here is that we're looking into using gas in Brazil, which again will improve our overall energy efficiency, the overall energy cost, and not least, it will also reduce our emissions and improve our climate footprint. The major new residue area investment is on budget. We spent roughly NOK 900 million, and we have about NOK 100 million to go, which mostly will be spent in 2017. We did commission the press filters this year in the third quarter, but there's still a little bit of work left to be done in the residue area, and it will be up and running by the mid of 2017. It's also important to say that this is not a pure maintenance project.
The new press filters is a new technology that's never really been used in a red mud area before. It also significantly reduces the environmental footprint, as we will be able to store about four times as much residue on the same square foot, acreage as we did with the old technology. In addition to the environmental footprint, it also, of course, then reduces the operating costs because we use less space, but it also improves our ability to recycle and recapture soda from the process and again, lower operating costs.
The operational improvements that we've just been through are clearly reflected in the cost position on a stand-alone basis but also compared to our peers. The right-hand side shows the implied alumina cost development for the combined B&A portfolio in the third quarter of 2016 compared to the same period last two years. As you can see here, the cost has come steadily down, supported of course by currency with a significantly weakening BRL versus the dollar. It's also very much supported by the improvements that we just been through. The implied margin in the third quarter of 2016 is at the same level of as in the third quarter of 2014. Remember that prices are almost $50 lower compared to 2014.
Improving our costs is of course important, but it is even more important for us to improve in relative terms. If you look at the CRU cash cost curve of 2016, it places Hydro and Alunorte very competitive, competitively in the low second quartile. Now, since cost curves are often imprecise and often quite dynamic, benchmarking to peers, as you know, can be quite complicated and challenging. But nevertheless, it gives us a confirmation, at least together with our internal figures, that we have a very competitive position in the B&A market. With that, let me invite Simon Storesund, Head of Commercial in B&A, to the stage.
Thank you, Eivind. Good afternoon, everyone. I will start to talk about the very interesting alumina market, which has gone up by $100 over the last few months. Then I will turn over to our commercial strategy and commercial optimization. As you can see here, the alumina market has gone up dramatically. Looking back two or three months, I have to say, and I have to admit, I guess, that I did not foresee the market going up towards these levels. We had a discussion in my team back at that time, and we said that maybe the market can go to 270, 280, but no more than 280. That's impossible. Now we are here at $326.
We had our Chinese trader, and he said that, "Well, the market may go higher, and it may stay longer at the higher levels." We have to listen to the Chinese. There have been some major factors driving this market. In my opinion, the three most important drivers has been the Chinese metal prices that has been higher. This has been driving the Chinese smelter restarts. In addition, the Chinese smelters have had to build inventory prior to the winter, and this is very important. They have to buy more than they really need. In addition to this, there has been several other factors also driving the market in the bullish direction.
This was mentioned also by Kathrine Fog, and that is the Chinese transportation constraints. That has been a big issue in China. Then you have also the higher raw material costs, as you can see here on the left-hand side. In addition, there has been a lack of imported material available from the West because of three refineries shutting down in the Atlantic. Looking forward, we are seeing now that there are numbers coming through from China, which indicates that the alumina production is increasing slightly more than the metal production. Of course, the supply-demand balance will determine the price, and I'm not allowed to talk about the price anyway because I have Eivind Kallevik here and Svein Richard Brandtzæg, so let's see what happens.
We believe that the floor has been lifted from the 220, 230 levels we saw. Over to the commercial strategy. In B&A Commercial, we have had the same pricing strategy since 2011, and that is to price our products on index. We still believe that that is the correct way of pricing our alumina based on its own fundamentals. As you can see from the graph, finally, Svein Richard Brandtzæg, we are up to 65% on index, and this has taken us 5 years of struggling and now we are there. It took time. Now we can, we can also enjoy it in the beginning of next year with high prices on alumina. We will realize 65% on the price you see today.
That's a good starting point. On the bauxite side, we are also there pricing our bauxite partly linked to the alumina index. We also get higher bauxite price these days from that. To the Atlantic alumina market, we have seen big changes in the market. 3 refineries shut down. As you can see from the graph here, we had a big discount in the Atlantic market going back two or three years. Now the market is basically flat, and it's balanced, and we expect that to continue for 2017. One of the key objectives for BNA Commercial is to create value above index. This is also what we call commercial optimization.
In order to try to explain this further, we have split it into four different areas, as you can see here. To start with the first one, this is really entering into the right contracts with the right customers at the right terms. We are also putting increased focus on the hydrate market, which I will revert to. We are also focusing on third-party sourcing to improve our optimization on the global portfolio. China. In May 2015, we established a Chinese trading entity in Beijing in order to get closer to the Chinese market and also to improve our margins. We have supply chain optimization. This is basically about geographical swaps, freight optimization, and to find the smartest logistical solution for our portfolio.
Here there is also money to be made for us. Reducing the merger cost was mentioned. Here you can see we have done, I would say good job in reducing it from $17 million to $5 million this year. A big reduction in cost. Finally, I just want to highlight two areas that I mentioned. The hydrate market. Hydrate is basically the same as alumina, but it has not gone through the final step of the production process, which is calcination. Otherwise it's the same. Hydrate is going to the so-called chemical grade market into product consumption areas such as you see here, it's water treatment, mobile phones, ceramics, et cetera. This is an interesting market that we will continue to focus on.
We have seen in 2016 opportunities opening up as two refineries shut down in the U.S. and the customers were looking for some new suppliers. There we have stepped in, and we have signed some new long-term contracts in the hydrate market. Then on China, the Chinese entity. Putting this entity in place has definitely given us a better handle on the market. We get the market information quicker. We are able to understand, I think, China a bit better. We have also seen that our margins are improving. We can now take advantage of the arbitrage between the Chinese price and the Western world price, from time to time there is a big difference.
We can also decide whether we want to ship from Australia and sell it straight away or ship it into our own inventory and sell it two or three months later. Also selling in smaller lots means that we have actually now selling into 3,000 tons, 5,000 tons lots instead of 30,000 tons, which means that we can reach more customers, and we can also increase our margins. All in all, I would say that the Chinese entity has been a success. We have definitely got closer to the market, and that is very important for us. I think that's it. Thank you very much.
To sum up, let me now update you on the progress we've made so far in 2016 on our midterm goals that we showed you last year. Very happy to say that we are green on all the key targets, whether it's operational, commercial or energy related. We have a safety statistic which is very much below the Hydro average, which is something that we are extremely proud of in BNA, and we are, as all entities in Hydro, aiming for an injury-free environment. As mentioned, we are well ahead on the better ambition and well ahead on track on our production targets for both Paragominas as well as Alunorte. We have shifted our PAX alumina exposure to 50% in 2016, and it will be 65% according to plan for 2017.
As I mentioned, well on the track to reach the one-to-one reforestation ambition for next year and to reach and to close the gap by 2020. The good progress shown here on this page will of course not mean that we will continue to strive for further improvements. We will always do that and set new and stretched targets. I think the development so far in 2016, and taking that into 2017, gives us a very strong foundation to deliver on the next milestones. Thank you. I would like to introduce Arvid Moss, Head of Energy, to the stage.
Thank you, Eivind, and good afternoon to all of you. I will give you some selected updates on what's going on in the business area energy and also update Hydro's strategy when it comes to climate and sustainability. First, on energy business area. We have what we call a dual mission. One is to own, operate, and maximize the value of Hydro's power assets in Norway. We have approximately 10 terawatt-hours of production in a normal year. The second mission that we have is to provide competitive power sourcing and global energy, that energy represents 50% of the cash cost from the bauxite mine to the smelter to the production of metal. To have competence in energy is extremely important for succeeding in this business.
We provide support to all the business areas globally, as I will refer to later. These two missions we try to fulfill every day in our business area.
2016 has been an active year on both these two missions. Let me start with the sourcing of wind power to Primary Metal in Norway. I think that's the first global contract, 20-year contract made to source a smelter with wind power. Very interesting concept. We also entered into a new power contract in Germany for the smelter in Rolled Products, and I'll revert to that. Brazil is very challenging from a framework condition point of view, and we see that the energy area is especially challenging, and we have stepped up our efforts there to support the business areas activities down there. Also we have increased our focus on our own small units, the Hydro Energia.
That to be in the market to gain learning and use that for the sourcing activity for the other business areas. This summer, the parliament passed the amendment to the law on industrial ownership to hydropower. That means that our opportunity, our solution space for the reversion has increased. We also, over the last two months, have opened two small power plants in Norway that were both put in place at schedule and at cost. We are, of course, supporting the total agenda of Hydro through better, bigger, greener. On better we are doing at utmost to optimize the value of our own assets in Norway.
I think one of the most important deliveries we have on better is actually on the energy contracts that we support to the other business areas, contract negotiations, et cetera. On bigger, yes, we try to get more out of the assets we have. We try to see if there are other opportunities at the fringe of these, but it's also here to get more out of the surplus situation in the Nordic to support Primary Metal primarily. On greener, I mean, we are in a way at the core of green in Hydro to develop hydropower and new renewable power is really important for the whole position of Hydro as a sustainable company. We try to promote this responsibility and have a green energy policy as something we promote globally.
Let me move to the power markets and an update on that. We have for many years heard about the Energiewende in Germany. We now see also that this is a global Energiewende. It is not something just happening in Germany. If you look to the graph to the bottom here, you see this is new installed generation capacity of power. You see that over the last years, the green column showing that investments and capacity in renewables is now larger than in fossil fuels. This is happening now with a higher speed than ever. We also see that new wind power and solar power, we see now projects that have a full cost at $30 per megawatt hour. This is really good news, seen in a global perspective and from a climate perspective.
This is very much supported by the very low interest rates that we have had over the last years. I guess in this room there are many views about low interest rates, but for sure it is good for this paradigm change because both solar and wind power is very CapEx intensive and low OpEx. That you really now in a period with low interest rates can put in place so much capacity has really taken down the full cost quite a lot. This is really now fueling the change to green economy. We see also to the right here that now we see the detachment between the GDP growth globally and the CO2 footprint. Things are moving fast now. Kathrine Fog also talked about the different sourcing patterns for power to the smelters in the different regions.
I think it's important to have in mind that there are certain global drivers that really influence the power price in all markets where there are some kind of free markets. What happened with the USA oil and gas has really contributed to a shift downwards, and it seems to stick. It's also clear that the coal balance in China has an influence globally. We also see that the interest rate levels, as I just mentioned, has an influence on the global cost picture for energy carriers. This translates in different ways into the power prices in the different regions. In Germany, it's very much the short-term marginal cost for a coal-fired power plant that sets the price. These global drivers come into play.
Short term, that is really driving the price, but also longer term, of course, the policy from the European Union, the targets for 2030 and 2025 and 2050 will influence the CO2 price, and that will of course influence also the power price. In Norway, on top of what is global and in Germany, we also have hydrology as a key parameter in the short term. But longer term, of course, also the transmission cables to the continent will have an impact. In Brazil, it is short term very much about hydrology and both short and long term, it's a lot about politics. It's a lot about politics and framework conditions. In U.S., it's really, in a way, that's where the source of some of this is coming from. What does it mean for the Nordic market? Can we explain what has happened?
Because to predict the future, you should be able at least to explain what happens in history. If you use the price modeling that most companies use, you can have quite a good bridge between the market price in 2008 and the price level in 2016, which is actually half of what it used to be. This is in real terms. The reduction in CO2 price and the reduction in the coal price each has contributed EUR 10 per megawatt hour to the reduced price, and then demand has had a small effect. That is really what explains the change. You can hear a lot in the Nordic market about certificates and data. That has destroyed the power market prices in Nordic. It's not.
It's the other global drivers and the European drivers that have really had the main impact. What does it mean going forward? I think last year we had some indication of what each of these parameters meant. I think if you look at the graph to the right there, you see the forward curve as it has been over the last years, going down from here two years ago, last year, the red, and now the yellow. Of course, here you have the introduction of more nuclear power in Finland, probably pressing the price down, while there is an expectation reflected in the forward curve of interconnectors coming into operation after 2020, as well as some effect on the energy carrier prices.
This is really something we follow very closely and will be interesting to see going forward. Moving on to energy and update on some selected topics. First, this one I showed last year was explaining some of the drivers for value creation in energy. I will just comment upon two of these. One is taxes. They will change the tax regime for power used in our own internal smelter system in Norway from this year. We expect that to come from the government soon. The other that I will share with you is the effect of changing some of the power contracts. This I would like to share with you is to explain the effect of the result effect internally of repricing of an internal agreement.
It will have some residual effect in energy, but also in rolled products from 2018. I think it's important to have in mind the historical price curve for the Nordic markets in 2000 until today. In 2008, at that time, was a high-priced environment. LME was around $3,000, and then also the power price was very high, and the forward price was at the level you can see there on the blue curve. At this time, we secured 250 MW or a little bit more than 2 TWh with the purpose to serve the Husnes smelter in Norway. At that time, that was a joint venture with Rio Tinto Alcan. That was a contract from 2013 to 2020.
Between Germany and Norway, and it was partly used for sourcing Rheinwerk for the period 2013 to 2017, i.e., for five years. That was done at a time when, and where you see the red line there, you see the forward curve at that time. Normally, all sourcing done by energy is done back to back with primary. But in this case, we as energy took over the majority of the power and market exposure, as not all contracted power had an industrial use at that time. We have had a P&L effect in energy over the last years of approximately -NOK 200 million related to the differences between the contract price entered into and the net sales price.
In 2015, we sourced power for the period 2018 and forward, and we priced the internally delivered volumes from 2018 to 2020 accordingly. That means the remaining part of the external contract. This will lead to an additional negative result of approximately NOK 250 million in energy, but a similar improvement then in Rolled Products. For Hydro, this is neutral. The blue curve there shows the forward curve when we enter the external contract in Germany, and you see here how this has dropped over the years. Of course, in 2021, we will be out of this external contract. That will, in its isolation, improve then the energy results with between NOK 400 million and NOK 500 million, and without any similar negative results for the other business areas.
That was a walkthrough of this special contract. Normally, we don't do it because we have some gains, some losses, but since there will be a substantial extra switch of result effect from 2018 and forward, we want to share this with you. Shortly on the update on the industrial ownership and the reversion in Norway. Last year we signaled that this was a change of law that we hoped will come true. Now it has come true, and I think that while the earlier legislation secured us the full value of the power after reversion. In addition now, we also have the potential to maintain the physical volumes, to use them into our smelter system, and not only as a financial owner.
We're working on this to see if we can establish then a larger entity where we have less than one-third. Short update on the balance between supply and demand for power in Norway and outside Norway. The one to the left shows the balance in Norway. The line is the consumption, 63 terawatt-hours. The blue is the captive power that we have. The gray until 2021 is the Statkraft contract, and the green are other contracts. We have entered then the Fosen contract this year, and to the right you see the global picture outside Norway, and there the purple is the Rheinberg contract that we just talked about.
All in all, we are moving forward in what I would call a moderate price environment, and we're working to continue the sourcing both in Norway and for selected smelters outside. We are working towards all the business areas on their energy strategy. We try to provide insight into the local markets to optimize the power portfolio, energy portfolio, and we lead the power sourcing negotiations. Securing of security of supply is also part of what we contribute with. In Brazil, over the last year, we have stepped up our presence, stepped up our work on framework conditions. Also, as Eivind Kallevik mentioned, we are very active when it comes to the fuel switch project to move to gas instead of oil.
I think on Primary Metal I've talked through what we have done, and on Rolled Products, we have now Rheinberg fully supplied until 2025 with power. All in all, we are on track with the targets that we put forward last year. Let me then finish off with the sustainability and climate agenda. I think that is something you have heard a lot about today through all the business areas, that this is very high on the agenda for Hydro. It is, though, some paradoxes that I would like to show you when it comes to the industry as such. If you look at the smelters, if you look at the emission from the smelters, the green part here is the direct emissions from the electrolysis, while the blue columns there is the emissions from the source you use.
If you use hydropower, of course, you have zero extra emissions, and if you use coal, like in China and India, you have 8 times as much emissions in total. You have 16 tons of CO2 per ton aluminum. Producing it in Norway, you have less than 2. Of course this is very important for the total footprints. If you then look at where has the increase come from, the green at the bottom, which is hydropower, has hardly increased over the last 15 years. Obviously, most of the growth has come in the gray area, which is coal-based. Of course this means that the total footprint globally of aluminum produced until the production phase has not moved in the right direction. This is something we as an industry will have to address. Why is that?
Because we know that governments will act. We know that after the Paris Agreement, they will address more and more these emissions. As many of the others have talked about, customers will react. Customers want aluminum with a lower footprint. This is an industry challenge that we have to address, but the industry is a very vague thing. In the end, it's about companies. How does companies act? As Svein Richard Brandtzæg has already shown, we have the industry's most ambitious climate strategy to be carbon neutral from a life cycle perspective by 2020. We have as put up very clear targets in all areas that we will deliver on. This is already, let's say we have been through all this today, and you see that this is now very much anchored in the business area strategies.
This is not just a staff work saying about we're going to be climate neutral. This is really about business, and that is what is so important. If we do not drive business in a climate-friendly way, we will not be sustainable. This curve here shows where we are heading. We are probably going to be carbon neutral sometime between 2018-2020, depending how fast we are moving on some of these things that are listed there. As you heard, of course, use of recycling or post-consumer scrap may vary with market conditions, so that's not everything we control, but we try to have enough levers so that we will reach this target by 2020. It, in the end, this is more than just climate. It's really the whole sustainability agenda that we need to drive.
Being a leading company is about setting the right targets for the future in this area. We believe that the sustainability agenda in broad will gain an importance. We are therefore one of the founders of this Aluminum Stewardship Initiative, where we have many other companies from producers to users, but also where the civil society is included. The point here is to establish guidelines and a framework for how to measure sustainability from the bauxite mine to the end use. I think that's the point that we are an integrated aluminum company makes it much more easy for us to really prove that we are operating in a sustainable way all the way through the value chain.
For me and for us in the management team, performance culture, yes, it's about cost, it's about enhancing market value of products, but I think that more and more we internalize the aspiration that we want to have also a performance culture when it comes to climate and sustainability in broad. We want to be attractive, a viable company also for the next hundred years, for our shareholder, shareholders, for our customers, for society, and for our employees. That is Hydro. Thank you.
Thank you, Arvid. Simon and Eivind, could you please join us on stage again for the last Q&A of the day? Again, please remember to state your name and company. Any questions? James?
Thanks. James Gurry from Credit Suisse. Just in relation to the bauxite and alumina division, I understand MRN's still probably got a pretty short mine life. Can you give us an update on what the ownership council might be thinking about extending that mine life sometime and your own security of supply for what you use that bauxite for? Given that you've just gone back to Brazil after a couple of years being out of the country, can you tell us what's changed, what's different? Do you guys feel more comfortable having capital invested there or investing new capital there given that you do have a couple of potential growth projects on the horizon?
Sure. I think if we start with the MRN question, the management and the board is currently discussing the project as such, and it goes through the different decision gates, which in Brazil we call FEL stages. Right now we're in FEL-2 stage, and then we'll see after the next board meeting whether we progress to FEL-3 or not. The project is up and running, but it hasn't been decided yet. It's in progress. I think for me to come back to Brazil, as you say, I have some years of comparisons to do. There is a very different structure in the plants, I would say, if you take the plants first. You know, we talk a lot about production systems in Hydro.
Sometimes it's hard to visualize unless you're actually there to see what's happening. I think if you go to Alunorte today, and if you'd been there in 2011, there's a very different level of engagement, and there's a very different level of process control and understanding. You take each of the different key parameters, you break them down to the lowest level of the operators. They talk to them during shift meetings. They talk to them, talk about them during shift handovers. Actually you have a very, very different control of the production system that you have today compared to where in 2011. From that perspective, much more comfortable than I was in 2010, 2011, 2012 or 2013.
Mm.
Denis Jacquet from Crédit Agricole, the metals mining team. One additional question on the bauxite in Brazil. You said that for Paragominas, the production is above the nameplate capacity now for a few quarters. What's today the estimated mine life still with Paragominas in the current production level?
We still have, I would say decades of, bauxite resources, in the Miltônia plateau where we do mine from. Today, we mine from the area that we call Miltônia Three, then we need to open up Miltônia Five, which will be the next area to move into. We still have sufficient resources in place for a long period of time.
Okay. That's linked with the increased CapEx.
Yes, it is.
Okay. Finally, sorry, coming back again on MRN. You said there is a sort of evergreen offtake contract that is in place. Basically the fact that discussion with Vale didn't go through is basically irrelevant because you have this offtake contract. What is the potential risk that Vale would try to sell their stake to one of the other shareholders or, you know, to a Chinese player or whoever?
It's a good question. Of course, this goes back to the way we set up the Vale transaction, or the acquisition from Vale back in 2010. Originally, the idea was to acquire also the 40%, but then given that there are pre-emption rights in the mine, we would then have to disclose that transaction to basically who's who in aluminum. Because there you have Rio, you had BHP at the time, CBA, Alcoa, and so on.
What we signed was an evergreen offtake agreement with a predetermined price formula. What is also part of that contract is if they were able to sell those shares to another party, the evergreen contract would follow. Yes, if they sold it to company Z, they would still have to oblige to the supply agreement.
Eirik Melle, Danske Bank. Two questions. You talked about the climate paradox and the power source for aluminum smelters. If governments will act and customers will act, do you see then a shift in the forecast that you have presented here? Also, I see that there's a spike in Hydro after 2020 sometime. Can you just comment on that? That's one question. The other one goes on the Røldal-Suldal and the new kind of industrial ownership. Can you say anything about the timeline, any deadlines or how we should think about a possible deal? Who will come in with 6 terawatt hours? What kind of benefits will they have from any agreement?
Would there be any cost related, et cetera? Thanks.
Okay. Let me start with the first one. I think that when we say that governments will act, customers act, it's. I think that will be a combination of push and pull. There will be gradually there will be more and more cost associated with CO2 emissions. China, yes, they will introduce an ETS system in some years. We don't believe it will be very high cost initially, but gradually they will tighten the screw. China needs to do something with their emissions and coal footprint. We think that from that angle, there will be a limit to growth when it comes to coal-based aluminum. I think. Also in China, we see some companies starting to work with hydropower plants that use. Kathrine showed some plants.
There can be some more projects in China also based on hydropower or even wind if they manage to. Customers will pull, and that means that you may see that there is an extra value attached to green metal. We think that is also a direction with our sustainability profile, where we have the full control of the value chain, that we can also position our metal in a way that over time it should be able to attract more value to it or attach more value to it. That's the thinking, that from push and pull, we will see that that forecast is not coming true. The exact increase in hydropower after 2020, I think you should forget about. When you ask about Røldal-Suldal, it's something I think we...
It's not much more to say now that we have sufficient time. It's end of 2022. There are numerous regional power companies in Norway. There are many combinations thinkable. I don't think there's anything more to speculate about. We will work on it when time is right.
Any further questions? Yes, Jonas?
Okay. It's Johannes here with Handelsbanken. Could you say a few words on the FX in Brazil and how that impacts your operations? I know you have talked about this before, but these things moves quite a lot, so if you can talk about that. And also on that subject, on the improvement program, you're ahead of schedule here. How much of FX has influenced that you're ahead of schedule? If I'm right, the Brazilian real is probably weaker now than a year ago. See if you can elaborate on that, please.
Sure. If you take the FX part first, in the short run, the OpEx, roughly 40% of that is BRL related, on the cost side, in the short run. Then remember that part, the fuel oil that we buy is BRL priced. Over time, we expect the fuel oil to move in tandem with the international fuel oil prices. The improvement programs, it's an easier answer, if you like. Because when we do these programs, what we say that we will measure only the things that we can control ourselves. Whether currency moves up or down, whether bauxite prices move up or down, we don't take the benefit nor the headwind of it. It's normalized for macro factors compared to when we set the baseline.
Okay. Last question from James.
Is there any reason in keeping the CAP project still there? Because you've had two refineries, I think, close down in the Atlantic market. The price is up. And this project, I don't know if it goes hand in hand with an expansion of Paragominas still or not, but it always seems to be mentioned every year, but pushed out further and further.
We still believe fundamentally that CAP is a good project if you look at this on a global scale in terms of energy efficiency, the location, and the bauxite quality we have in Paragominas. It is, as we've said also last year and the year before that, the market is not there and the prices doesn't sustain it, and supply-demand balance is not there. Again, as we've said before, when the time is right, it is still a good project.
Good. That concludes this Q&A session. Svein Richard, if you could, join us on stage for some concluding remarks.
Thank you, very much. I'm not going to repeat all the messages you have got today. We have been through almost 170 pages of information. But I think you will now understand why we will continue to develop Hydro as integrated value chain with all the opportunities that we see along the value chain of the company. And also, of course, that we are very motivated to continue to improve the profitability of the company with several improvement programs in place. We will do that within the framework of a strong financial situation, a strong balance sheet that keeps the financial flexibility going forward. With that, thank you very much for joining us, and please look at the last minutes of a movie here. Thank you.