Thank you. Good morning, and welcome to Hydro's Q4 presentation and conference call. We will start with the presentation, followed by a Q&A session. Both will be hosted by our CFO, Pål Kildemo, who will be presenting Q4 and Q2 going forward, while our CEO will also present at Q1 and Q3, which will include more of a strategic focus. The presentation slides we will walk through can be seen on the webcast. The link to the webcast, as well as the slides, can be found on hydro.com. Please note that you will need to dial into the conference call to be able to ask questions at the end. It will not be possible to ask questions over the webcast. If there are any media inquiries for one-on-ones with Pål after the presentation, please contact Head of Media, Halvor Molland.
With that, I turn the microphone over to you, Pål.
Good morning, and welcome from me as well. I hope you and yours are keeping well and healthy in these unprecedented times. Before getting into the quarter, I would like to briefly let you know of a revision to our business area names. To reflect our new strategic direction, we have simplified and updated the names of our business areas to better reflect the business. For instance, Primary Metal is now Aluminium Metal, reflecting that the business area holds both primary and recycled metal. The other business areas are Bauxite & Alumina, Energy, Rolling, and Extrusions. Now, let's get into the quarter, where we are pleased to report that we have exceeded our 2020 goal for the improvement program and again, similar to Q3, generated a strong cash flow. Move to slide two, please.
Underlying EBIT for the fourth quarter was NOK 1.449 million, up from NOK 560 million in the same quarter last year, while our free cash flow also increased from last year, ending up at NOK 3.4 billion for the quarter, reflecting our focus on cash generation. In the fourth quarter, we were pleased to see an overall continuation of the economic recovery following the contractions caused by COVID-19 earlier in the year. Aluminium demand, especially in China, remained strong, resulting in a year-end surplus that was lower than what was originally forecasted. These fundamentals have resulted in a strong LME development during the fourth quarter and also a continued recovery in the downstream markets, where the year-over-year volume declines each quarter have continued to be less and less through the year.
Within our downstream businesses, both Extrusions and Rolling recorded volumes which exceeded the market outlook for Q4, with Extrusions growing 70% compared to Q4 2019 and Rolling staying flat year-over-year. This has been supported by recovery in key downstream segments like automotive through the second half of 2020. We are also pleased to announce that we have achieved and even exceeded our improvement target for 2020, recording NOK 4.2 billion in savings compared to our NOK 4.1 billion target. This is particularly satisfying as we were slightly behind our target in Q3 due to the pipeline maintenance in Bauxite & Alumina. Another positive development this quarter has been the Lyse Kraft transaction. On December 31st, 2020, Hydro's RSK hydropower asset was merged with part of Lyse Produksjon to form a joint hydropower company.
The new company, Lyse Kraft DA, of which we now own 25.6%, secures us access to renewable hydropower for our aluminum production and ensures that RSK does not revert to state ownership in 2022. I will cover some of the details of this transaction later in the presentation. Finally, we are also pleased to announce the proposed dividend of NOK 1.25 per share for 2020, which will be paid in 2021 subject to approval from the annual general meeting in May. Our ability to pay dividends after a difficult and volatile year is due to the measures undertaking to support our strong cash flow generation. We safeguard our balance sheet and then service our shareholders before evaluating return-seeking investments in operations. We strive to provide our shareholders a predictable dividend, and we are pleased to continue delivering according to policy.
We aim to lift performance and cash returns to shareholders over the cycle and have as such revised our dividend policy to pay a minimum of 50% of underlying net income over the cycle with a continuation of the NOK 1.25 per share dividend flow. Let's now move into some of the key figures for the fourth quarter. Next slide, please. Slide three. If we start with EBITA, then underlying EBITA increased by 24% compared to the fourth quarter last year and rose by 3% compared to Q3 2020. On free cash flow, I am very pleased that we are able to deliver improvements both from last year but also from the last quarter, which was particularly strong. Our strict focus on working capital and capital expenditures, in addition to ongoing improvement efforts, have been key drivers in generating this cash flow.
Our underlying RoaCE for the last 12 months rolling at Q4 2020 remains around 4%, in line with Q3. If we just look at the underlying RoaCE for Q4, we would have come in over 7%. If we use both assumptions that we currently see, we are actually around 10%. If we move on to the business area key indicators, upstream, we have experienced a slightly higher cost base compared to the previous quarter in B&A, as there have been some costs related to extraordinary alumina sorting due to the pipeline maintenance, and also some costs related to repairing a ship unloader crane used for unloading bauxite at the port we use in Parag. In Primary, we see a relatively flat development quarter-over-quarter, but a considerable improvement in cost position compared to last year.
Downstream, we see increased volumes compared to both Q4 2019 and Q3 2020, and in general, performance is somewhat better than the underlying markets. It is also worth noting that Q4 2019 was relatively weaker than expected, which exaggerates our improvement this quarter somewhat year-over-year. We exceeded our 2020 improvement target and are so far on track to deliver on the 2021 goal of NOK 6 billion in total compared to the baseline of 2018. Please move over to slide four. The global aluminum market ended 2020 in a surplus of around 3 million tons. This surplus is considerably lower than the earlier forecast this year, which were close to 5 million tons. The global market was fairly balanced in the fourth quarter and stable from the third quarter due to a strong recovery in demand throughout the year.
However, 2021 will still see a surplus in the aluminum balance driven by strong production rebound in China, but also the rest of the world due to consistently high prices since the summer of 2020. In addition, announced smelter closure candidates did not close after all, which keeps production levels at higher levels than expected. It is also interesting to note that external consultancies providing balance estimates are in relative close consensus for 2020 run around 1.5-2 million tons, following a 2020 where there were large variances between each company outlook. We will get back to the specific demand drivers and expected developments later in the presentation. Please change to slide five. Let me now get into a high-level results overview. Comparing to the same quarter last year, Q4 underlying EBIT is up by around NOK 800 million.
On the positive side, we have experienced lower raw material costs than one year ago. This is mainly driven by lower alumina prices for Aluminium Metal, but also supported by lower carbon prices, as well as somewhat lower bauxite and caustic prices in Bauxite & Alumina. We also saw a currency gain as the NOK and BRL weakened against the dollar. In addition, stronger margins and volumes downstream drive our downstream results, supported by the cost improvement efforts in both Extrusions and Rolling. These positive developments were offset primarily by additional other costs, which includes around NOK 500 million in maintenance-related costs in Bauxite & Alumina, and also around NOK 250 million in antidumping duties and provisions in rolling. Please move to slide six. If we compare the fourth quarter results to the third quarter, we see a stable development quarter-over-quarter.
Upstream, the positive impact from prices is coming primarily from Aluminium Metal, where we have had both higher LME and premiums from Q3 to Q4, driving around NOK 1.1 billion of the gain. On the other hand, the volume uplift comes mostly from Bauxite & Alumina, where we managed to produce around 90% of nameplate in Q4 and reached nameplate by year-end. We also saw an improved result in Energy as production remained relatively high and prices rebounded following lows in Q2 and Q3 due to more exports and normalization of weather patterns. These positive developments were more or less offset by higher raw material cost upstream, with both alumina and power costs rising in Aluminium Metal and higher fuel oil costs in Bauxite & Alumina.
These costs are in addition to the aforementioned maintenance-related costs in Bauxite & Alumina and antidumping duties in rolling, impacting our other costs here by around NOK 700 million. The additional NOK 200 million in other costs results from cyber attack insurance compensation booked in the third quarter of 2020 of around NOK 200 million. Please move to slide seven. Our underlying EBIT for 2020 of NOK 6.1 billion increased compared to 2019 by NOK 2.7 billion. The main contributor of our improved results was lower raw material costs, primarily in Aluminium Metal, which had lower prices on alumina, energy, and carbon in 2020 compared to 2019. Positive currency effects from weaker BRL and NOK to dollar also contributed around NOK 4.5 billion to the annual underlying EBIT.
Finally, the ramp-up of Alunorte, which produced at 87% of nameplate capacity in 2020 and reached nameplate by year-end, was the main positive contributor in terms of upstream volume uplift. It is also satisfying to see that we managed to reduce NOK 1.7 billion in fixed costs in 2020. Over 80% of that saving comes from our downstream businesses, Extrusions and Rolling, and is a true credit to how these organizations managed to exhibit cost discipline amidst the volume and margin pressure our industry experienced and continues to experience due to COVID-19, represented by the NOK 2.4 billion negative impact in downstream volume and margins. In addition, our realized alumina prices fell 18% year-over-year to $268 per ton, and our realized aluminum price fell 7% from 2019 to 2020. Resulting in a combined hit on the underlying EBIT of around NOK 7.5 billion. Please move to slide eight.
I would also like to highlight the positive contribution of the improvement program in 2020. Without this improvement and the cost discipline shown by our business, our underlying results for 2020 would have fallen to under NOK 2 billion. Our uplift in volumes, primarily in Bauxite & Alumina, provided a lift of around NOK 2.6 billion. The containment efforts in Albras also contributed positively here. Our cost out initiatives and other improvements, both downstream and in Aluminium Metal, added another NOK 1.6 billion in savings. I will now go to the next slide where we will take a closer look at this program in 2020 and moving forward. Please move to slide nine. Our original improvement target for 2020 was an accumulated saving of NOK 4.1 billion. During Q3, we mentioned that due to the pipeline maintenance, we were slightly behind this goal.
We are pleased to note that we not only reached the overall goal for 2020 but exceeded it by around NOK 100 million. The main driver of the NOK 4.1 billion goal has been ramping up our volumes at Alunorte from the embargo levels of 2018. During 2020, we produced an 87% of nameplate capacity, and we expect Alunorte to continue running at nameplate capacity in 2021 after having reached this level in December. It is delivering on cost improvements ahead of schedule like in rolling, which enabled us to reach this goal. In rolling, for example, our original target was NOK 150 million. Strong momentum behind organizational rightsizing and procurement initiatives throughout 2020 have provided a saving that is over three times higher at NOK 500 million.
At the Capital Markets Day 2020, we announced that the improvement program would be extended to 2025 and include an even more ambitious cost-saving target of NOK 8.5 billion. This program is front-loaded and after 2020, we are nearly halfway there. In 2021, we plan to secure an additional NOK 1.8 billion in savings driven by further cost out, but also volume gain, like producing at nameplate capacity in B&A and the ramp-up of Husnes Line B. We will also continue to work toward our commercial uplift ambition of NOK 2 billion, which comes on top of the NOK 8.5 billion through market and customer-driven opportunities within Aluminium Metal, Rolling, and Extrusions. Capturing more volumes and margins through our greener products, Hydro CIRCAL and Hydro REDUXA, is one important lever towards this commercial ambition. Please move to slide 10.
If we take a look at the key financials for the quarter, revenues were stable at around NOK 36 billion for the fourth quarter. Underlying EBIT came in at NOK 1.4 billion, as explained on the previous slide, it is worth noting that a NOK 5.3 billion gain on the Lyse transaction is held outside of underlying EBIT, I will explain this in detail later. With depreciation of around NOK 2 billion, underlying EBITA amounted to NOK 3.5 billion. Financial income of NOK 1.4 billion for the fourth quarter included a net foreign exchange gain of NOK 1.5 billion. This primarily reflects a stronger NOK versus Euro, affecting the embedded derivatives in Norwegian power contracts and other liabilities denominated in Euro. Our tax expense amounted to NOK 849 million, or about 10% of income before tax.
The relatively low tax rate was positively impacted by the tax-exempted gain of NOK 5.3 billion related to the Lyse transaction. Overall, including the transaction gain, this provides a net income of NOK 7.3 billion, up from negative NOK 0.7 billion in the same quarter last year. Underlying net income was positive NOK 0.8 billion compared to negative NOK 0.3 billion last year in Q4. Consequently, underlying earnings per share was NOK 0.35 per share, up from a negative NOK 0.12 per share in Q4 2019. Please move to slide 11. Now, let's review items excluded during Q4, which total around NOK 5.3 billion. As usual, we excluded some timing effects in Q4 of around NOK 100 million. These are more or less offset by impairments related to the ongoing restructuring in Extrusions.
The main item excluded this quarter is the NOK 5.3 billion in transaction-related effects related to the completion of our deal with Lyse. This is an accounting gain based on the fair value estimate of our ownership stake in Lyse Kraft DA, representing the depreciated asset base in Energy. Please move to slide 12. Underlying EBIT for Bauxite & Alumina increased from a loss of NOK 75 million in Q4 2019 to NOK 116 million profit in Q4 2020.
The quarter saw positive effects from currency and lower raw material prices. The BRL's weakening against dollar contributed around NOK 700 million, and in addition to upside from lower bauxite and caustic prices, though these raw material gains were partially offset by higher fuel oil prices. In sum, these factors created an overall lower implied alumina cost per ton in Q4 2020 of $241, which is $14 lower than last year's level.
Overall margins were up by around NOK 5 year-on-year. These positive effects were partially offset by the lower prices, additional sourcing costs due to pipeline maintenance of around NOK 280 million, and costs associated with crane repair work of around NOK 170 million. The production at Paragominas resumed after extended pipeline maintenance in October, and alumina production at Alunorte averaged around 90% of nameplate capacity, as we guided in the last quarter. During the fourth quarter, Hydro undertook extraordinary repair work on the crane, which is used for unloading bauxite from ships. The repair work caused additional operational costs but did not have a material impact on production or shipments from the Alunorte refinery. As part of our continuous asset integrity program in B&A, we will continue preemptive maintenance where necessary to ensure the robustness of our operations.
By doing so, we should be able to avoid major production curtailment, and in the long run, it will help us to increase the life cycle of our assets and reduce risk and cost. Lower realized alumina prices by $9 per ton also negatively impacted the quarterly results by a negative NOK 120 million compared to the fourth quarter in 2019. If we look into Q1, Alunorte production is expected to be around nameplate capacity. However, we do expect the costs associated with the crane repair to impact Q1 results also by around NOK 200 million, but we do not expect extraordinary sourcing costs as experienced in the fourth quarter. The crane incident in Bauxite & Alumina is notified to the insurance companies.
In addition, we expect somewhat lower raw material prices in comparison to Q4, given today's market prices, while current alumina prices indicate somewhat higher realized alumina prices than achieved in the fourth quarter. Please move to slide 13. Underlying EBIT for Aluminium Metal increased from NOK 155 million in Q4 2019 to NOK 844 million in Q4 2020. In general, we experienced stronger margins year-on-year. All-in prices were marginally higher, but the main contributor was a lower cost position per ton, mainly driven by lower alumina prices, lower carbon prices, and lower fixed cost and depreciation, which accounted for a positive impact of NOK 600 million year-on-year. From Q3 to Q4, we managed to mitigate some of the fixed cost increases, which were anticipated for Q4.
The improved margins quarter-on-quarter come mainly from higher LME and premiums, which lifted underlying EBIT by nearly NOK 1.1 billion from the third to fourth quarter. When it comes to the outlook for the first quarter, we have by the end of December, sold approximately 65% of our primary aluminum production forward at a price level of around NOK 1,980 per ton. This includes pricing effects from our strategic hedging program, which I will get back to a bit later.
On the premium side, we have secured around 69% at around NOK 273 per ton, and we expect a premium level within the range of NOK 225- NOK 275 per ton. When it comes to the cost side, we are expecting somewhat higher fixed costs compared to Q4 2020 due to seasonality of policy credits, the timing of some property tax invoices, and the further ramp-up of the Husnes smelter.
We are also expecting some higher power prices from a new internal power contract and currency effects, and I will get back to that a bit later. Please move to slide 14. This quarter, Aluminium Metal delivered an underlying EBIT of NOK 248 million , compared to NOK 132 million in Q4 last year. The results increased compared to the same quarter last year due to higher results from the sourcing and trading activities and positive currency effects.
Underlying EBIT from our recycling facilities were unchanged, however, included a positive insurance refund in Q4 2019, implying stronger results excluding this effect. Higher sales volumes were the main driver of those. Excluding the currency and inventory valuation effects, primarily currency, the result for the quarter was NOK 257 million, which is up from NOK 184 million in Q4 2019. Looking into the next quarter, our recycling facilities are operating at normal capacity levels.
As always, remember that trading results and currency effects in metal markets are by nature volatile. Please move to slide 15. The market for our downstream business areas is on the path to recovery. Within Rolling, we see a flat development in our sales volumes from Q4 2019 to Q4 2020, which was a better performance than the overall market, which was down 2% in Europe for Q4. The overall rolling market outperformed the external analyst forecast, provided at Q3, of a decline of 5% in Europe. Rolling continues to optimize its portfolio by increasing the share of higher margin can and auto, and in addition, this shift will further support our strategy of higher levels of recycled volumes. For Q1, we see indicators that the rolling market is continuing its recovery, and external analysts estimate 2% growth in Europe for Q1 compared to Q1 2020.
Our Q1 outlook for Rolling is more or less in line with the market. We do have some planned revamps for casting centers and production lines in Q1. For the full year 2021, external analysts expect the market to improve 8% in Europe as volumes are recovering from the COVID outbreak of 2020. Please move to slide 16. The results in Rolled Products decreased to a loss of NOK 188 million in Q4 2020, compared to NOK 34 million profit in Q4 2019. The result from the rolling mills decreased mainly due to a NOK 259 million impact of the imposed U.S. antidumping duties in October 2020. This negative impact includes a provision of NOK 60 million for Q1.
The negative effects were partly offset by lower costs from the ongoing improvement program mentioned earlier in this presentation, where Rolling has delivered over three times higher than its original saving target, achieving NOK 500 million in savings through 2020, primarily through organizational restructuring, metal cost optimization, and procurement initiatives in logistics, maintenance, and direct materials. In addition, the quarter saw some positive currency effects. The Norsk smelter results increased, driven by lower raw material costs.
If we look into Q1, the antidumping duties is expected to have a negative impact of around NOK 100 million. For 2021, we have only one contract of some 1,000 tons in total, which we are working with our customers to try and replace. The U.S. International Trade Commission is to decide whether there was harm to U.S. producers by April. If no harm is found, the duties could disappear and potentially even be reimbursed.
We believe that the duty is unfounded and unreasonable. We will work with all related parties to minimize the effects of this and will keep you up to date on any developments. The strategic review continues with the aim to evaluate the best ownership set up for Rolling within or outside Hydro. The result of the review will be announced in due time. Please move to slide 17. In Extrusions, we saw a 7% increase in our sales volumes compared to Q4 last year, 12% higher in Europe and 4% higher in North America. The improved volumes exceeded the market, which was flat in North America and down 3% in Europe. We should keep in mind that Q4 2019 was a challenging quarter for us.
The market development was better than what was expected three months ago, when it was predicted declines of 8% and 7% in European and North American markets respectively. In the automotive segment, we saw overall solid recovery continue into Q4. OEMs are increasing production again, and there were less supply disruptions during second wave of COVID-19 than during the first wave. Some of the demand rebound is linked to restocking following destocking in Q2. Nevertheless, underlying demand is also improving, which is evident in increasing auto sales, especially electrical vehicles, where we see strong growth in Germany, and also growth in the premium segment, where aluminum is the most exposed material. When it comes to the first quarter outlook, external market expectations are indicating further improvement. External analysts expect an increase from same quarter last year of 7% and 2% in North America and Europe, respectively.
In total, we expect the development in line with external market in the first quarter. If we look at the full year 2021 forecast, external analysts estimate a growth of 9% and 10% in North America and Europe respectively. We believe, based on our internal estimates, that we will be roughly in line with the rest of the market. However, there is still uncertainty in the early part of 2021. Now, let's take a closer look at the financial results for Extrusions. Please move to slide 18. Underlying EBIT for Extruded Solutions significantly increased from NOK 85 million in Q4 2019 to NOK 511 million in Q4 2020. The results were positively impacted by increasing volumes also from Q3 of 5% and 7% from Q4 20 19. The increase in volumes from Q4 2019 and previous quarter was driven by a general recovery in most segments.
Comparing the Q4 results to the previous quarter, we see, as mentioned, a positive impact of slightly higher volume, while fixed costs increased due to seasonality related to regular maintenance towards year end. In addition, the third quarter was positively impacted by cyber insurance payments. Continuing the trend from previous quarters, Q4 results were also positively impacted by reduced costs from the ongoing improvement and restructuring efforts, as well as other temporary cost measures implemented amid weaker demand that are expected to start returning with increased activity levels. We are also pleased that Paul Warton is now in place as the new EVP for Extrusions to continue lifting profitability and cash flow from Extrusions in the years to come.
If we look into the first quarter, Extrusions is working hard to support their earnings with the ongoing portfolio optimization, fixed cost reduction initiatives, and procurement optimization. Extrusions continue to firmly control their cost development into the first quarter, with the objective to retain a maximum of savings as the market slowly resumes from the COVID-19 turmoil. Please move to slide 19. Underlying EBIT for Energy increased from NOK 296 million in the fourth quarter 2019 to NOK 352 million in the fourth quarter of 2020. The quarter saw a significant drop in prices compared to Q4 2019, mainly attributed to a strong hydrological balance, with prices averaging NOK 137 per megawatt hour, compared to NOK 392 per megawatt hour in Q4 last year.
If you compare prices to the previous quarter, prices have increased from low levels of just NOK 52 per megawatt hour due to more export and normalization of weather patterns. In January, we have seen drier and colder weather, nearly eliminating surpluses in the hydrological balance, so things can change very fast. We have seen some improvements in power prices in the Nordic region, with average NO2 spot prices of NOK 499 MWh so far in January. The uncertainty is still large and will depend on the weather going forward. Please let me also remind you that NO2 spot prices are publicly available, and the realized price levels for the company should not deviate significantly from the prices observed on the Nordic power exchange for NO2. Keep in mind that the prices and production can change fairly quickly in response to hydrological development.
We have quite some other elements impacting the first quarter results, which I will refer to on the next slide. Please move to slide 20. This is a page I would like to spend a bit on time on to update on the changes as a result of the Lyse/RSK transaction and some other changes in Energy to keep in mind for modeling going forward. As mentioned at the start of the presentation, we announced the merging of our RSK hydropower asset with part of Lyse's production to form a joint hydropower company in October, and the transaction closed on December 31st. We are very pleased to see the conclusion from the discussion in the Standing Committee on Energy and the Environment on the parliamentary question related to the conversion of the RSK concession.
Establishment of Lyse Kraft DA builds on the principles established by the Stoltenberg II government in 2008, that was further developed by the Solberg I government, and secures both public ownership to the RSK hydropower plant and long-term predictability for industrial activities and value creation. It is comforting to see that the continuation of this framework is supported by such a broad coalition, comprising the parties in the government and the main two opposition parties, Arbeiderpartiet and Fremskrittspartiet. The Lyse RSK transaction is reported as a sale in Q4, the proceeds of the sale is the value of Hydro's 25.6 ownership in Lyse Kraft DA. As part of the fair value assessment and purchase price allocation, an external third party has estimated the fair value of our share of Lyse Kraft DA to NOK 7.8 billion , corresponding to an EV multiple of approximately NOK 3.3 per kilowatt hour.
Hydro applies IFRS accounting policies, which include elimination of Hydro's proportional share of unrealized profits from transactions with associates and joint ventures. Elimination of 25.6% of gross gain represents NOK 1.8 billion. This creates a recognition of NOK 5.3 billion accounting gain after adjustment for carrying book values for RSK and IFRS elimination. Additionally, Lyse Kraft DA will be owned as an associate and accounted for as an equity accounted investment, as opposed to RSK previously being held as a subsidiary. Due to this reclassification, the pro-rata share of Lyse Kraft DA's net income will be added to Energy's EBITDA, as opposed to previously, when RSK's EBITDA was directly reflected in Energy's. This will reduce Energy's EBITDA by NOK 100 million-NOK 200 million annually.
It is important also to note here that our own production volumes will be lower by around 750 GWh as a result of the transaction. This, however, is partly offset by higher quality assets, representing approximately 300 GWh, of which circa 200 GWh impact EBITDA and around 100 GWh is related to higher tax shield, resulting in lower tax payments. The volume loss related on EBITDA, you could estimate to around NOK 130 million, assuming a price of NOK 300 per megawatt hour and cost of NOK 70 per megawatt hour, and correspondingly to around NOK 30 million gain on net income, resulting in a net effect of -NOK 100 million for Hydro. Over the medium to longer- term, we expect increased positive synergies from the transaction to mitigate some of this loss.
In addition to Lyse Kraft DA, there are a few other changes in Energy to keep in mind for 2021. A loss-producing supply contract, which had a negative effect of around NOK 650 million in 2020, has now expired. The large loss in 2020 was impacted by low spot prices as part of the contracted spot exposed, and in addition, significantly impacted by weaker NOK compared to Euro, as this was a Euro-denominated one. The second contractual effect to keep in mind from 2021 in Energy is related to the new internal contract with Aluminium Metal. Here we see a positive effect of around NOK 750 million compared to the internal contract portfolio from 2020. The positive effect here compared to using 2019 and 2018 as a starting point, is a stronger Euro resulting in a currency gain in Energy.
It is important to note that the gain on the internal contract will be partly offset by the expiry of external sourcing contracts in Aluminium Metal, but still resulting in a net increase in power costs in Aluminium Metal of around NOK 300 million-NOK 350 million compared to 2020. These effects depend on the currency rates used, and these calculations were based on a Euro/NOK rate of around 10.8. It is important to keep in mind that we are ramping up our battery and renewable growth units, and while it is difficult to give full guidance on the cost level for these areas at this stage, one could assume a preliminary estimate of additional costs of between NOK 100 million-NOK 200 million related to operational costs, but also the project development which will go into their respective portfolios. Please move to slide 21.
Other eliminations was -NOK 435 million in Q4 compared to -NOK 67 in Q4 last year and +NOK 213 in Q3. Other is mainly comprised of head office costs and costs related to holding companies. This quarter we had NOK 300 million in costs compared to NOK 223 last year and NOK 204 in the third quarter. This quarter's eliminations amount to - NOK 135 million, reflecting higher internal volumes. Please move to slide 22. We move now from the business areas to the development in net debt. Overall, our net debt position decreased by NOK 2.1 billion. We started Q4 with NOK 9.9 billion in net debt. We generated underlying EBITDA of NOK 3.5 billion, and we then had a release of working capital of around NOK 1.6 billion. We had originally forecasted a build of working capital for the period.
We ended Q4 with lower inventories and higher payables, mainly in the downstream business area, in addition to normal seasonality effects. Other operating cash flow adjustments of +NOK 0.4 billion includes dividends from equity accounted investment, reversal of mark-to-market effects being offset by interest, taxes, and some other negative effects. We generated net cash flow from operations of +NOK 5.4 billion in Q4. Investments came in at around NOK 1.9 billion, and we also paid out the 2019 dividend in November of Q4, totaling NOK 2.6 billion. The net effect of exchange rates on our debt and cash also lowered overall net debt by around NOK 1 billion. At the end of Q4, we ended with NOK 7.8 billion in net debt, our lowest level since 2018. Please move to slide 23.
If we look at the adjusted net debt at the end of the fourth quarter 2020, that has decreased by around NOK 3.5 billion compared to Q3. Net debt decreased by NOK 2. billion, as I just explained, and net pension liabilities decreased by NOK 1.6 billion due to higher discount rates in Norway and positive returns on Norwegian pension plan assets, slightly offsetting the lower discount rates in Germany. Other adjustments were up NOK 0.3 billion. With that, the total adjusted net debt, including equity accounted investments at the end of Q4 amounted to NOK 28 billion, down from NOK 31.6 billion at Q3 2020. Please move to slide 24. Let me then move on to some comments on our financial policy and hedging. We see that the aluminium industry is a cyclical industry, where the periods with above cost of capital returns are of limited duration.
As a result, Hydro has reviewed the situations where the use of derivatives could make sense in order to support the group's financial targets and is more actively evaluating the use of such derivatives, although still to a limited extent, and with the majority of our exposure still remaining unhedged for the upside. In line with our hedging policy, we have the flexibility to hedge LME or currency in certain cases, and we have normally done this with respect to transactions or planned startups. On our current profitability agenda, we are constantly looking into making our portfolio more robust and delivering on our return ambitions over the cycle. Hedging is one lever which can be used to support the strategic agenda by reducing volatility in earnings and improving downside risk scenarios.
Last quarter, we mentioned a forward contract for 30% of B&A to BRL/ dollar exposure to 2021 and 2022. The aim here is to reduce the volatility and uncertainty in Alunorte's cash flows and ensure our cost curve position for Alunorte remains robust as it is currently. In addition, we have secured LME hedges for 250,000 tons of primary aluminum or 10% of our volumes, each for 2021, 2022, and 2023. The hedges have locked in LME prices seen in late December, early January. You can see that the guidance for Q1 volumes includes the effects of these hedges. Parts of the corresponding raw material exposure has also been partially secured through further derivatives and physical contracting, locking in an attractive margin for parts of our volumes.
Please move to slide 25. For 2020, Hydro's Board of Directors proposed a dividend of NOK 1.25 per share, reflecting Hydro's robust financial situation and taking into account the demanding year caused by the COVID outbreak and the general volatility of the aluminum industry. The dividend represents a payout of around NOK 2.6 billion, which is to be paid in May, subject to approval by the annual general meeting of shareholders. The proposed payment demonstrates our commitment to shareholders, providing a competitive shareholder return compared to similar companies.
It also represents a dividend yield well ahead of the aluminum peer group when comparing dividends paid out during 2020 compared to the year-end 2020 share price. On average, our five-year payout ratio based on reported earnings is 65%, which is above our over the cycle dividend policy, which was to have a 40% payout ratio with NOK 1.25 considered as a floor.
Moving forward from 2021, the board has increased the payout ratio in our dividend policy to minimum 60% of underlying net income with a floor of still NOK 1.25 per share, reflecting our ambitions to lift performance and the resulting cash returns to shareholders over the cycle. Given the current market price environment, we're off to quite a good start. Please move to slide 28. I will now conclude with an update on our capital return dashboard for 2020, summarizing our performance on key financial targets and priorities for the year. Starting at the balance sheet and the key ratio of funds from operations to adjusted net debt, we have seen 39% in the last 12 months, which is an improvement compared to 35% last quarter and nearly in line with our target of about 40% over the cycle.
Following our strong second half, free cash flow for the year is up NOK 7.7 billion, NOK 6.6 billion of which came in the second half of the year, and supported by an overall release of NOK 2.4 billion in net operating capital during 2020. We have also maintained our CapEx cuts in 2020 to preserve liquidity. We aim to cut CapEx to around NOK 7 billion-NOK 7.5 billion, and we ended up undertaking NOK 6.4 billion in investments in 2020. However, NOK 700 million will be carried over to the 2021 CapEx budget. Thus, we update our outlook for 2021, which was originally NOK 9 billion-NOK 9.5 billion, to NOK 9.5 billion-NOK 10 billion. Our longer- term outlook or medium-term outlook for 2022 to 2025 remains at NOK 9 billion-NOK 9.5 billion.
On that, I would like to thank you for joining the presentation and invite you to stay for a Q&A session, which will begin shortly. Thank you.
Thank you, Pål. Operator, we're now ready for questions. Thank you.
Thank you. Dear participant, if you'd like to ask a question, please signal by pressing star one on your telephone keypad. If you're using a speakerphone, please make sure your mute function is turned off to allow your signal to reach our equipment. A voice prompt on the phone line will indicate when your line is open. Please state your name before posing your question. We'll take our first question from our participant. Your line is open. Please go ahead.
Hi. Good morning, Pål. It's Jason Fairclough from Bank of America. Two quick ones for me. First, through the COVID downturn, you were selling LME grade products to traders. I'm just wondering if you can confirm that this is finished. Second, and probably related, we are hearing of shortages of certain sorts of products, and I guess particularly can stock. Any similar product shortages that you're seeing across your product suite?
Thank you, Jason. Good questions. If we start with the shortages, the latter ones, then we see across the system that the demand recovery is stronger than most players had expected, both when it comes to can, automotive, and building and construction. As you see from our operating capital figures, these are at lower levels than we thought, and that is much driven by reductions in inventory beyond the levels that we normally keep as the minimum stock levels. We are able to supply most of our customers, but we are actually having to move some deliveries a bit out in time because demand is so strong. If we had to have more rolling flats, for example, available in the system, then we could be able to increase the production to even more than what we are doing today. It is not causing an issue for us.
It is a good trend. Ideally, we would be able to ship out more and not have to delay it a bit into next year. The first question was on?
Sorry. You were selling commodity grade products to traders.
We have seen our value-add premium. Exactly. We have seen the value-add premium products sales increase. We're selling less standard ingots typically to traders or to others to finance. As you know, part of our operations are on a normalized level, also producing standard ingots, and those are being continuously sold to different market participants. We are moving there and have moved the value-add product production more towards the normal levels for our portfolio. Linking it a bit back to the comment on sheet ingot, we also see the same picture on extrusion ingot. We are producing and selling at capacity and running everything at full speed, and also ramping up Husnes as fast as we can.
Compared to earlier quarters when we spoke, when a lot of this was driven also by restocking, we see more signs now of this being supported by fundamental demand figures also. A bit more sustainable demand based on what we're delivering, but still uncertainty in the months following January and outwards.
Well, just if I could follow- up. If we look at other commodities, what we're seeing is that the premiums or the spot prices are rallying quite hard as lead times extend. Do you feel like you're fully taking advantage of this tight market environment in terms of premiums and pricing?
I think as you will have seen from Aluminium Metal results, that premiums moved up. They moved up to the higher part of our guided range. Also a bit above market expectations based on what we saw in Q3. With these increased volumes, we are also seeing increases in premiums. You're seeing increases driven by demand. You're seeing increases in premium as driven by more regulatory matters. For example, like in the U.S., where the Midwest premium is now up at record high levels. We are able to benefit from that, as you will also see into our Q1 guidance. We will continue to ensure that we work for increasing those premiums as long as the market tightness continues.
As you're also aware of, LME prices are and have been trading quite a bit above the 90th percentile on the cost curve, reflecting at least the tightness we're seeing in the market now.
Okay. Thank you very much, sir.
Thank you, Jason.
We'll take our next question from next participant. Please go ahead.
Good morning. It's Liam Fitzpatrick from Deutsche Bank. Two or three questions from me. Firstly, on Rolled Products. Even without the duties, it's pretty close to breakeven, so it seems like it could have a better owner elsewhere. Can you give any more color on, at least the timing on when you think you will have made a decision either way to keep or sell? Second question on CO2 compensation for the Norwegian smelters. In terms of timing, when do you think we get more clarity on the numbers for 2021 and beyond? Final, hopefully a quick one on alumina costs. If we ignore the crane or the maintenance impact, in terms of the net effect into Q1 versus Q4, do you expect overall costs to be down just given the lower sourcing costs? Thank you.
If we start with a question on Rolled or Rolling, as it is now called. If we just start with the underlying results, on an EBIT level, excluding the antidumping duties, we are still around breakeven levels. It's important to remember that this also includes effects of the cleanup and the restructuring which are ongoing, represented by, for example, higher depreciation related to curtailment of certain parts of the operations. The EBITDA development in a year impacted by quite a significant COVID-related effect has been very much undersupported by the NOK 500 million in improvements delivered by Einar and his organization. If we look at the strategic review, like earlier sessions, we will announce to the market when that is completed, and we hope that that is in not the too distant future.
On CO2 compensations, I unfortunately can't answer on behalf of the Norwegian authorities. The ball is in their court now, and they need to make a decision before the annual budget for 2022. Anywhere between now and the end of the summer is the timing that they will tend to do this, is our expectation. Looking into both sides of alumina and development into the first quarter, of course, the fact that we don't have to source extraordinary alumina equipment to supply our customers on their contracts will have quite a large net on net effect, close to NOK 300 million. If we look at the other elements and market prices, then fuel oil is a bit up. Coal is flat to somewhat up, and caustic soda is also remaining weak. There's not a lot of other cost push apart from crude oil.
The totality should still be a good positive development in cost for the first quarter, even if you include the NOK 200 million in ship unloading.
Okay. That's clear. Thanks, Pål. Thank you.
Thank you.
We'll take our next question from next participant. Please go ahead.
Good morning. Jatinder from Exane BNP Paribas. A couple of questions, Pål. Just one on hedging. Again, is 10% meant to reflect any sort of level which represents only your external volumes from Hydro Aluminium, or is there any other thinking behind 10%? Is there potential to change that number to a higher level? Second question on dividends. Good to see minimum payout being lifted on the floor of NOK 1.25. Is there any scenario where management or board could think about lifting that floor as well? Thank you.
Thank you, Jatinder. We start on the hedging, the absolute percentage does not reflect our external position or similar. What we are looking at is we want to remain majority exposed to the movement in the margin. Given the cyclicality, and given the absolute integrated margin that we're seeing now, utilizing some of the better years to safeguard the cash flow for potential lower cyclical years is something which we believe over the cycle should make sense. 10% is what we've committed to now as we've been building up the strategy and program, ensuring that we have the right system in place. This could increase somewhat in the coming periods, if we see that we are able to sufficiently safeguard margins, but not significantly above the levels that you are seeing today.
When it comes to the dividends, we have the floor, as you are aware of course, that provides a form of predictability and has safeguarded a certain yield over the cycle. Being a cyclical exposed company, we would rather reward shareholders in times where the cash flow is available for doing that, instead of committing to large payments in periods where we are on the cyclical low side. We have seen during COVID, which is maybe an exceptional circumstance, how big that dividend payment can be of the total balance sheet outlook and the review from rating agencies. As we see it today, given the current cyclicality in the portfolio, we would rather keep the floor at this level and give extra back when earnings improve.
Over time, if we move forward with the strategic direction and some more diversification in earnings, we might think differently about that. That is not currently something we're evaluating today.
Good. Very clear, Pål. Thank you.
Thank you, Jatinder.
We'll take our next question from next participant. Please go ahead.
Yeah, morning, Pål. It's Amos Fletcher from Barclays here. Just a couple of questions. I just wanted to ask around. Firstly, the NOK 1.8 billion incremental cost reduction in 2021 that you're targeting, can you split that out by business area? The second question was just around working capital. Is there a risk that you have to rebuild working capital during the course of 2021, given your comments around inventories being run down? The final question is just also around there's headlines about the collective lawsuit that's being brought against you in the Netherlands. Is there anything you can sort of say about that in terms of potential impacts, provisions, et cetera? Thanks.
Thank you, Amos. Good questions. If I start backwards and moving to the top, when it comes to the Cainquiama lawsuit, then our entities received a notification about a Dutch lawsuit, and we will respond as requested before the Dutch court. However, it is worth noting that the matters brought forward are already being discussed before Brazilian courts and Brazilian authorities, and that the Cainquiama Association has since 2017 filed five lawsuits in Brazil against different Hydro entities in Brazil. We are looking into the case now, and we'll follow up as we should. On the other elements, the working capital, there is an expectation from our side. A large part of what has been released this year should come back into 2021 if we see markets remaining where we see them today, and the volumes coming back as the external analysts expect.
As you saw from our report, there is an expectation of quite a decent growth in the year-on-year volumes in Extrusions and Rolling that will drive some operating capital. We are ramping up the Husnes smelter and completing the ramp-up of Albras. That will drive some operating capital, then you have the price effect on top. Yes, already into the first quarter, you could typically see a build above seasonality on operating capital. If we look at the improvement program and the amounts that you talked about for next year, then around NOK 1 billion of these are improvements in the business areas and around NOK 900 million is reflecting a ramp-up in B&A volumes. Since we're not running at nameplate, but 87% for the year as a whole, you still have a ramp-up effect into next year.
Of the NOK 1 billion in improvements, that is spread quite well among the business areas, With a larger share in Extrusion, Rolling, and Aluminium Metal and less in B&A. B&A primarily has the volume effect into next year.
Okay. That's great. Thank you.
Thank you, Amos.
We'll take our next question from next participant. Please go ahead.
Yes. Hi there. This is Ioannis Masvoulas from Morgan Stanley. Most of my questions have been answered, but just have a couple for you. The first one is, in terms of the Primary Metal or Hydro Aluminium business, what sort of cost progression should we be expecting across the various moving parts sequentially for Q1? Secondly, given the new accounting changes in the Energy division, what sort of depreciation base should we be assuming for the group going forward? Thank you.
If we start with the cost development in Primary Metal, then the Energy cost is the biggest moving element there into the first quarter, as you saw from the Energy slide, that the higher prices on power contract internally has a negative effect in Aluminium Metal, which is now, compared to 2020, not compensated by other high-cost expiring contracts. Depending a bit on what currency rate you use and the starting point, you could see between NOK 100 million and NOK 200 million in increased Energy costs for Aluminium Metal Q4 versus Q1. That is before you potentially take in anything on the CO2 compensation side. The other elements move with the market. Alumina prices are somewhat up compared to what we saw in the fourth quarter. If prices remain there, that could be around $30 or so per ton.
Carbon costs are also somewhat up, and if prices remain where we see in the market now, that could also be around $30 or so per ton. You have fixed cost, which also typically increases a bit into the first quarter. If you look at depreciation, the Lyse Kraft transaction per se shouldn't really be impacting depreciation to a large extent. As you know, the RSK assets were so written down that they didn't add a lot to the depreciation in the existing system. When we write up these assets, this is the increased effect that we're talking about, which is between NOK 100 million-NOK 200 million for the year as a whole, which will not hit depreciation, but it will hit the EBITDA line as we account Lyse Kraft DA as an equity accounted investment.
The overall depreciation line should be marginally reduced by that transaction, but then the implicit depreciation effect is higher, but it will not hit the depreciation line. We can spend some time with IR to go through the spreadsheets and the details if you want afterwards.
Understood. Thanks very much.
There are no further questions at this time. Dear speaker, please go ahead.
Great. Thank you everyone for joining us today, and please don't hesitate to contact us if you have any questions afterwards. Thank you very much, and have a good day.
Thanks.