Gunhild Grieve - IR Markus Krebber - CFO and Member of the Executive Board

Vincent Gilles - Credit Suisse John Musk - RBC Ahmed Farman - Jefferies Lueder Schumacher - SocGen Sofia Savvantidou - Exane Alberto Gandolfi - Goldman Sachs Michel Debs - Citigroup Mark Lewis - Barclays Ingo Becker - Kepler Cheuvreux Nick Ashworth - Morgan Stanley Vincent Gilles - Crédit Suisse Deepa Venkateswaran - Bernstein

Good morning to everyone on the phone and to those who are joining us via webcast. I'm joined here by Markus Krebber for the presentation on the First Half of Fiscal Year 2017. As with our Q1 presentation, we will concentrate on RWE standalone. Where appropriate, we have shown the relevant group figures, which you can also find on the interim report. In order to focus on your questions, we have kept it short. So let me hand over to Markus.

Thank you, Gunhild, and good morning, ladies and gentlemen. We had a good start to the fiscal year after the reorganization of RWE. This not only relates to our operating business, but also in particular to the positive developments that have occurred in the field of our nuclear power business. Let's turn to the revenue of the first half of 2017 on slide 3. The operational performance of the RWE Group was quite positive. EBITDA improved by 7% and adjusted net income even by 35%. The latter was supported by a strong improvement of the financial result. This development is in line with our expectations, and hence, we confirm the outlook we have given for 2017. For RWE's operating businesses, the normalization of the earnings in our trading business was the main driver for the positive development. European Power also performed well due to a higher contribution from our commercial asset optimization in the first 6 months of 2017. Based on the H1 results and our current expectations for the rest of the year, we can improve the outlook for the European Power division. For the group's KPIs, we can confirm the outlook for the year as a whole and expect to achieve a result at the upper end of our guidance. On June 7, the German Constitutional Court published its decision that the nuclear fuel tax was not compliant with the provisions of the German constitutional law and declared it null and void, retroactively. We have already received the refund for the paid tax of approximately €1.7 billion and expect another approximately €250 million of interest by year-end. With regards to our 2017 income tax, we expect an effect of around €200 million. The reimbursement of the tax amount is accounted for in the non-operating result and the effects on the interest result and taxes are adjusted for in the reconciliation of adjusted net income. We plan to pay our shareholders a one-off special dividend of €1 per share with the predominant part of the reimbursement used to improve the company's financial strengths. We believe this is an appropriate use of proceeds against the backdrop of a 2-year dividend holiday on the one hand, and we aim to strengthen the balance sheet on the other. In the context of nuclear energy, we can also report the successful conclusion to the restructuring of responsibilities for nuclear waste disposal. On June 16, the new law entered into force, according to which the government is now in charge of the handling and financing of the interim and final storage of nuclear waste. On July 3, we transferred the relevant funds to the government. For RWE, this was €7 billion, including E.ON's minority stake in our Emsland nuclear plant. The economic stake for us was €6.8 billion. What is important in this context is the fact that we concluded a public contract with the German government. This contract provides us with a higher level of legal certainty related to the release from the liability and also establishes the conditions for transferring the radioactive waste to the federal authorities. Furthermore, it forms a basis for ending numerous legal disputes involving nuclear energy-related issues between the utilities and the government. The only major outstanding topic in the context of nuclear energy is the settlement around the constitutional claim to shorten the operational lifespan of German nuclear power plants. In this context, the Federal Constitutional Court decided in December last year that some aspects were unconstitutional and that the utilities were entitled to compensation for damages. This refers to generation volumes which cannot be used as well as a loss in value of investments which the plant operator had made on the basis of the lifetime extension set out by law in 2010. The legislator was given until mid-2018 to formulate suitable regulations for this compensation. We estimate that our claim will amount to a low triple-digit million euro amount. That's it on the nuclear topics. As part of their regular rating review, the 3 leading agencies confirmed their credit rating for RWE. All agencies rate our creditworthiness in the investment-grade category. Regarding the outlook, Moody's and Fitch rates the outlook from negative to stable, which corresponds to the outlook by S&P. Let's then move to the operational performance. Slide 4 shows the development of our adjusted EBITDA. For RWE standalone, it amounted to €1.4 billion. The dominant driver in H1 was the recovery of our trading activities to a normalized level. The adjusted EBITDA also includes the energy dividend of €683 million for fiscal year 2016 that we received in April. In the previous year, we had a comparable figure of €730 million, which represent the pro forma appropriation of profits of innogy subsidiaries before the IPO. Hence, this figure was higher as it did not include any minorities from the listing of innogy. Slide 5 provides the detail of the performance of the Lignite & Nuclear division in H1 and the outlook for this division for the full year. Adjusted EBITDA has come down mainly due to declining realized generation margin. This was partly offset by the phaseout of the nuclear fuel tax in 2016. Last year, we had a burden of approximately €100 million from the tax in the first half of the year. Furthermore, after the decision to sell our Hungarian lignite operator, Mátra, we have classified our participation as an asset held for sale. As a consequence, the earnings contribution is recorded in the non-operating result from Q2 2017 onwards. In the last full fiscal year, Mátra contributed €60 million to adjusted EBITDA and €35 million to adjusted EBIT. In this context, we also undertook a revaluation at our Mátra participation, leading to a value adjustment of approximately €300 million, which is reported in the non-operating result. Minorities carry roughly half of this amount. The decline in generation margins are also the main reason for the outlook for the full year of significantly below previous year. We now also expect an additional negative earnings impact of a medium double-digit million euro amount due to some unplanned outages at our Neurath lignite blocks F and G. On a positive note, we are well on track with our efficiency improvements. The European Power division showed a strong earnings decline, as we had significant positive one-offs in 2016; among others, from the sale of real estate in the UK. Excluding nonrecurring items, EBITDA was up year-on-year. This is due to a strong earnings contribution from our commercial asset optimization. From today's point of view, we can improve the outlook for the division to significantly above 2016. This is mainly driven by the positive earnings trend from our commercial asset optimization. Furthermore, the book gain from the expected sale of Littlebrook real estate in H2 will contribute to higher earnings. This brings me to our current hedge positions on Slide 7. Compared to our situation at the end of March, we almost closed the implicit fuel hedge for 2018 and reduced it slightly for '19 and '20. Our hedge prices for 2019 and 2020 have improved by €1 per megawatt hour compared to the end of March. As a result of our implicit fuel hedge strategy, we could benefit from the improvement of the fuel spreads, which you can see on Page 8. Now onto the performance of the Supply & Trading division on Slide 9. By far, the biggest lever is the return to normalized earnings in our trading business after the losses we incurred last year. Looking at the full year 2017 and confirmed by the results of the first half, we are optimistic we will return to a normalized EBITDA level as well, which we see on average in the order of €2,000 million -- €200 million per annum. Ladies and gentlemen, Slide 10 provides the earnings drivers down to adjusted net income. Our clean net income numbers reached €883 million in H1 2017. This is a strong improvement in Q2, driven by the recovery of our trading business and the fact that we have received the full innogy dividend of €683 million in Q2. Besides the typical adjustment of the nonoperating result, we have adjustments in the financial result, the minorities and tax position. The financial result we have adjusted for the effects from the changes in discount rates for nuclear and other long-term provisions as well as the interest component on the nuclear fuel tax. In the noncontrolling interest position, we adjust for the effect of the write-down of our Mátra participation. Taxes were adjusted accordingly. Details of the reconciliation can be found in the appendix on Slide 15. Now onto our distributable cash flow on Slide 11. The first six months of the year distributable cash flow reached €636 million. This is disproportionately high and may not be extrapolated for the full year, mainly due to the inclusion of the full innogy dividend in H1 and the typical pattern of our operating working capital. With regard to the change in provisions and other noncash items, I would like to point to the timing effects of CO2 provisions mentioned on the slide. With the transfer of the CO2 certificates to the national clearing authorities in Q2, we have completed this year's utilization of provisions, whereas the addition are only halfway through the year. For the year as a whole, we can confirm our expectation for the change in provisions in the order of €650 million, as we have guided previously. The change in working capital is partly driven by seasonal effects in trade accounts payable and inventories, which are typically higher over the winter months. On a full year basis, this is usually more or less neutral. For the 2017 full year, we expect the change in working capital to be negative due to the phaseout of some working capital optimization measures. Given the current financing conditions and our liquidity situation, some of these measures are no longer economically favorable. The details of net debt are shown on slide 12. At the end of June, it stood at €4.3 billion for RWE stand-alone, approximately €2.6 billion below the end of fiscal year 2016. There are 3 main drivers: first, the positive distributable cash flow; second, the refund of the nuclear fuel tax of €1.7 billion; and thirdly, a positive effect from a reduction in provisions, mainly pension provisions, which accounted for an improvement of €450 million. One reason is that the discount rate for our German pension provision increased from 1.7% to 1.9%. Furthermore, pension assets performed well, and thus contributed to a decline in pension provisions. On slide 13, you can see the details of our outlook for RWE standalone 2017. I have already explained the reasons why we are much more optimistic for our European Power division and have upgraded the expectation to significantly above previous year. However, our earnings expectations for the Lignite & Nuclear division has slightly deteriorated compared to Q1 for the reason which I explained earlier. Overall, we can confirm the RWE stand-alone earnings KPIs. Providing that the second half of the year develops in line with our expectations, we are now optimistic to end at the upper end of the range. The same applies for the outlook for the consolidated group that you can find in the appendix on slide 16. This concludes my remarks, and we are now happy to take your questions. Gunhild, please.

Thank you, Markus. Although everybody knows it quite well, just as always remind you of our two-question rules. Operator, please?

Thank you. The first question is from Vincent Gilles, Credit Suisse.

Two very simple questions. The first one is on the non-operating results restructuring, €1.4 billion, and you got, of which €1.7 billion is the cash from the nuclear fuel tax. Obviously, where does the difference exactly come from? I understand Mátra will be there, but if you can help us bridge very precisely. And the second question is on the asset optimization in European Power. I'm trying to work out whether this improvement is only for H1, whether the improvement you saw during H1 can be extended into H2. And if you could help us quantify, all things being equal, what this improvement is. Is it €20 million, is it €40 million, is it €100 million? It would be fantastically helpful.

Yes, no problem. First, to your very detailed question, I mean, the answer is as simple. I mean, the major missing effect is Mátra, where we said the valuation adjustment was around €300 million. And then there are some minor, but offsetting effects, but very minor ones. So the two relevant factors are the reimbursement of the nuclear fuel tax of €1.7 billion; and Mátra, minus €0.3 billion. On commercial asset optimization in the European Power segment, I mean, the overall profitability contribution of commercial asset optimization is around €2 to €3 per megawatt hour as we have said and explained in detail at the Capital Markets Day. This is more relevant for the European Power segment than it is for Lignite & Nuclear, but please understand that we will not give more details how much exactly it is per segment. Now coming to your question how much was improvement, so the result from commercial asset optimization was around medium-digit million number higher than we had expected in H1.

You mean, double digit?

Double digit, yes, medium double digit. And I mean, our current expectation is that, that is a basis for our guidance, that we can maintain this improvement we made in H1. But H2 profitability will be on expected levels, so on the normalized lower level.

Thank you. The next question is from John Musk, RBC.

Two questions from me as well. Firstly, in the actual interim report, you talk a little bit about changing regulations on airborne pollutants and you seem relatively relaxed that these may be watered down when pushed into national levels. But if they weren't, what is the scale of the risk in terms of your assets and potential closures in Germany? And then secondly, with the -- obviously, the improved balance sheet post the nuclear fuel tax rebate, how should we be thinking about you looking at tuck-in acquisitions in European generation?

First one, on the breadth, I mean, you are right that we are relaxed because we have ranges and a very staged implementation path and it is now up to the national regulators. And as the German government already voted against the decision taken on European level because they thought it was not appropriate given which technology are available, but we can live with the ranges. That is our expectation that the German government will take a very cautious approach how to implement it. But now starting speculation, if they go for mid of the range or the lower range or another implementation phase, that's much too early. But your impression, how we look at it, is right. On the balance sheet side, yes, I mean, the nuclear fuel tax has brought down net debt significantly. Please keep in mind when we talk about the €1.7 billion, that we have promised €1 extra dividend, which equals to €650 million. So more or less, it's €1 billion. We are currently looking at all options, how to strengthen our financial situation that is on the debt side but also measures maybe to improve operating cash flow in the future, but it's too early to speculate. It will take some time, but we will have a very cautious approach here. Now the last one on M&A speculation, I mean, usually, we don't comment. I can just reiterate what we said before. Some portfolio optimization that we go for more diversified portfolio could make sense, but please understand that we don't speculate what it could be.

Our next question is from Ahmed Farman, Jefferies.

So just two questions from my side. I just want to make sure that I understood your comments around the changes to the basis of the guidance earlier correctly. I think on the negative side, you mentioned the additional outages on the lignite, which, if I understood correctly, was a double-digit million impact. Was the reclassification of the Hungarian power plant had an effect as well? And then, on the positive side, I think you mentioned a double-digit effect from the commercial asset optimization. But I think in the report, you also talk about better load factors and margins on the gas-fired feed. Is that a separate effect or is that included in your, the comments you made earlier? And then, my second question is your, on your energy Supply & Trading business. I think in the second half, you had a negative EBITDA in that business, sorry, in the second quarter. Could you give a little bit more color on that? And is that something that we should expect for the second half as well?

Thank you. Yes, let me maybe spend some more explanation on the guidance. First one, for the Lignite & Nuclear segment, it is correct, as you said, that the additional burden from the unplanned outage is a double, medium double-digit million euro amount for the full year, I mean, most of it in H2. And from the Hungarian operations, which are from Q2 onwards, not reported in the segmental results but in our non-operating result, we can only say because it's speculation when we sell and how much it would have been, we can only reiterate what we said about what the contribution was last year. So like-for-like, it was a €60 million EBITDA and around €30 million, €35 million EBIT contribution for the full year. But that is also a reason why the expectation for the full year is now slightly below what we expected end of last quarter. Now these are the 2 facts in Lignite & Nuclear. Now coming to European Power, right. In the interim result, we talk about different load factors for and margins on the coal and gas-fired power generation. More or less, these are offsetting factors. So we made more money than previously expected in gas-fired power generation but less in coal-fired power generation. The simple reason is that gas is more times price setting; the load factors for gas are higher given the very low spot gas prices we currently see. So gas is more often in the money than previously assumed. But these factors are more or less equaling out. So the real net effect is from commercial asset optimization and there I said before how much the improvement was. Another effect is the sale of the Littlebrook real estate, which also will contribute to the full year result. But already, on an operating basis, we expect a higher operating result. With the Littlebrook sale, it will drive the results significantly above previous year's results. And please keep in mind also in last year, in 2016, we had positive one-off effects from the sale of real estate in the European Power division. Last question on Supply & Trading. Yes, you are right. On a net-net basis after cost and -- or the typical volatile valuation effects, the contribution was slightly negative in Q2, but the gross margin contributed by the trading division was positive. And as you can see from the confidence we have about the full year, there are no indications from Q3 so far that we will not achieve a normalized profit level of €200 million or above.

Thank you. The next question is from Lueder Schumacher, SocGen.

First question is on Neurath. I think if I heard had you correctly, you said that the unplanned outage for first unit F and G, that will be the BoAs. What exactly is the reason for the unplanned outage and how long is it expected to last? That's the first one. And the second one is on your fuel hedges. I didn't quite get what Markus was saying as he was speeding up there, but I think you said that 2018 has now been closed, 2019 is still open. What is your general view? I mean, '18 clean darks, depending on what efficiency you use, still don't look that attractive. Do you think '18 is not fairly valued and there's more upside for clean dark spreads for the outlook curve?

On Neurath, yes. So what was the situation at Neurath? So the Block F was down from June, as you can also see from the remit publication, was down from June 2nd to July 21st. The reason was damages with the turbine blades. The loss of production during that time was around 1.2 terawatt hours, and the maximum capacity is also down now around 15 megawatt until the next regular maintenance stop. And we will see -- that was it on Block F. Since Block G is more or less exactly the same type, we will have an extended 17 days outages to also look at that in order to apply what we learned from the incident at F also to G. And so overall, we already lost production. The additional costs for the work we have to do at the turbine will amount in total to the double-digit -- medium double-digit million amount we mentioned. And on hedging, yes, I mean, sorry for being too fast there, but what I said is that we have closed our implicit fuel hedge for 2018. So now it is more or less a full hedge of the outright position. I mean, of course, we can start speculation what we think about spreads or at the absolute level of profitability here. As you can see on the next slide, 8, that the blended spread, so not -- it's not the clean dark spread here but the blended spread incorporating all assumed pass-through factors of the different fuel types and how often they are price setting. So in the end, that is the summary of our own model. There, you can see that calendar year '18 spreads are quite high and already above the level we have seen in the outturn for '16 and '17. That is also the reason why we have closed the implicit fuel hedge position. In the longer run, I mean, we have two effects on the spread side since power prices, absolute prior prices are not that relevant. The first one is, what is the type that's in the market? I mean, we can lots of speculate about it but what we will see is that conventional generation will calm down, the capacity in Europe, especially in Germany, over time, especially on the nuclear side. And the other one, this is more difficult to predict, what is the exact price consolation on the fuel side? Because that will determine which technology is in the money, which technology is price setting, what the pass-through factors are, how much carbon you need and so on. I mean, for us, the most relevant picture is this blended spread we show here, we have shared with you, and there, you can see '18 is already quite rich.

Is that okay, Lueder?

Yes, it's very clear.

The next question is from Deepa Venkateswaran, Bernstein.

So my 2 questions, first one is about your other, consolidation or holding. So just wondering what is the outlook for the full year because I think on the Q1 call, you said, I think you indicated roughly around €100 million. I don't see that impact coming through in H1. In fact, it's a positive contribution. So just wanted to see where that lands. And my second question is on the CapEx for the year. Whereabouts do you expect that to land? And again, how does that compare to where you see CapEx going forward?

Yes. Thank you, Deepa. On others and consolidation, you are right. H1 was a very positive figure, especially Q2 was driven by the dividend we received from our Amprion participation, which is shown in others and consolidation. We have also now much more clarity on the cost level. We see under others and consolidation, it was a very tricky exercise when we splitted energy and RWE. So we have now much more clarity on the RWE side and you are right. We don't expect €100 million. It's more now a medium double-digit million amount we expect for this year and also the years going forward, yes. On the CapEx side, we expect the pattern where we see more CapEx in the second half of the year, and we still stick to our guidance, that's going to be around €350 million for this year and also the years to come; and this is purely maintenance CapEx and some CapEx for the conversion of the Dutch plants into biomass.

The next question is from Sofia Savvantidou, Exane.

A couple from me as well. Looks like there have been quite a lot of sort of small changes in your portfolio, selling some assets, deciding to delay the mothballing of phasers, putting others on held for sale. I was wondering, are you seeing some opportunities or changes in the operating environment in terms of how you evaluate your coal versus your gas portfolio for now and for the next few years? Or is there just opportunistic adjustments that you've made that we've seen over the last quarter or so? And then, my second question is whether we could get an update on your views around your contract with Uniper on Datteln 4. We heard from Uniper's side that they are fully confident that the contract will be honored and/or the plant will be commissioned next year, which I would assume on your side that, that means you would have to be buying those volumes at that price. What is the, well, one, if you can give us an expectation on the court case; and two, what is the financial impact from you from that contract starting?

The first one on portfolio piece, don't read too much into what we have currently done. So it is a purely opportunistical and economically driven decisions we have taken. I mean, you are referring to Moerdijk 2 where we decided not to mothball it in 2018 but to at least run it into '19. And we will reassess our entire portfolio given what we see on the forward markets, so what the profitability level of this plant is. So this part is purely opportunistically, so don't read too much into it. On the broader context, we always said if we go for some portfolio measures, then we will go for more diversification of our portfolio to bring it more in balance, because we all know that it's very coal heavy. On the Datteln plant, yes, I mean, I can just repeat what I said earlier. It is -- for us, it's also a very clear case. We contractually agreed to take 450 megawatt out of the 1,050. The commission is delayed -- the commissioning is delayed by years and markets have changed significantly. So power today is a different product than what it has been seven years ago since we all know that you can't store it. We have made use out of an extraordinary termination right. Uniper has disputed that and now the decision is going to be taken in court. But even this court decision will not be the end because if you don't have an extraordinary termination right, you still can ask for an amendment of the contract because you are now living in a different world. So let's see what the first court decision will be and then we decide where to take it from there. We think we have a very, very strong case already for the first court decision to come. Speculating on the financial impact, I think it's much too early. We have made our assessments and we need to reflect our assessment in our books. But I can -- I mean, you will understand that we will not give any transparency on that.

Thank you. The next question is from Alberto Gandolfi, Goldman Sachs.

A couple on my side as well, if I may. The first one is more on the industry in general. And could you please elaborate on your outlook for carbon and particularly what are you doing in terms of hedging and policies there, considering that we are entering in the -- on the height of the debate on the MSR and there has been ongoing talks about a carbon floor? So if you can maybe elaborate your framework around carbon would be great. And the second one is just for peace of mind or peace of heart. Can you remind us your stance on innogy? So still a financial stake, still a value maximization strategy. And would you be open to swap part of this stake for a portfolio of assets?

Alberto, the easier question is the second one on innogy. Nothing has changed, okay? It's still a financial investment. We are exploring all options. We don't comment and let's see, yes. And I'm not speculating whether we're going to swap it for A, B, C or D. Purely value maximization, that is our approach here. And on the industry and carbon, yes, let me answer the question with 2 aspects. There was some speculation early summer and spring, that after the election of Mr. Macron we might see initiatives here. I think a very senior politician from Macron's party is a spokesman for energy policy, had made a couple of days ago a very strong statement and some of you have picked it up for the research report that Germany will not support national carbon floor measures. They support the ETS. That's it on the carbon floor side. We as a company also think that the ETS is the system where the European Union has exactly what they need in order to steer how much carbon will be, how much carbon emission we're going to see. Let's wait for the outcome of the Trilogue negotiations after the summer break where they will fix, hopefully soon, what exactly the parameters. In our view, it would not make a huge difference for which model they go. We expect a moderate impact on carbon pricing for the years to come. Our hedging policy so far, we have hedged the entire implicit financial carbon position we have until 2020, now almost 2021 already, but we are currently hesitant to go beyond that tenor because the road costs are very high liquidity; beyond '22 is very low. And of course, we want to see the outcome of the Trilogue negotiations to have more clarity where we think carbon prices are going. And after that is fixed, we might extend our hedging beyond 2021 already now. But that's going to be decided after we have the outcome. So I mean, if you want my summary, it's more or less, on the carbon side now, business as usual on European level.

The next question is from Michel Debs, Citigroup.

I have 2 questions. The first one is about your thermal fleet. There is a divergence between baseload forward curve and peak-load forward curves, especially in Germany with the peak load being much higher and much more in contango. So my question is, looking at your fleet, how should we think about this in terms of sensitivity? Can you give us the financial sensitivity to a change in the peak versus the base or maybe tell us what is the portion of the fleet that can respond as the peak load versus the portion of the fleet that will have to stay on the base contract? My second question has to do with the German election. Besides the fact that there have been talks about changes to CO2 legislation, is there anything for you to play for? Do you have anything to gain or to lose on energy policy on the German level from the outcome of the election in September?

Yes, thanks. Let me ask with, start with the second question because that relates to the question from previous speaker. The first, I mean, on the carbon side, I think, I mean, after the clear statement from the CDU party, that is a European topic. So we don't expect the German government to, other than on European, [indiscernible] on European level, they will do something on national level. The question that needs to be answered by the next German government is, what will they do in order to ensure security of supply. The next, during the term of the next government, we're going to see the phaseout of the entire remaining nuclear fleet. And that brings us, as we have discussed previously, in very tight market situations, and something has to happen because with the addition of renewables, load factors for thermal generation will come down; profitability will deteriorate further. So if nothing happens, we're going to see more and more firm capacity leaving the market. And we think it's not -- in the end, it will not be a question of money, because transferring the UK capacity market to Germany will maybe cost €2 billion. And it will already save a lot of cost fee, where the system suffers today from stabilization measures. So the net figure will be less than €2 billion. Compare that to the €25 billion for EEG every year that is not financially a relevant figure to ensure security of supply. It is too early to speculate. First, we need to see the results from the election, and then the forming of the new government. And then, we can discuss how that probably can look like. Your other question is also a very relevant one. I will not give you much transparency what it means for our fleet. But it will -- I mean, both baseload and peak-load spreads will go into our blended fuel spread where we take our hedging decision. So we run our model and we know exactly which plant will run at peak load. And so how often is in the money, that determines the position we have. And of course, we use both products, base and peak, to hedge our positions. So we will gain from higher peak load prices and the higher peak load prices are also incorporate into the blended fuel spread we have shown on Page 8. And since we are not willing to give you too much details because that would reveal our assumption on pass-through factors and so on and so forth, which we think are confidential information, but the relevant picture to look at is how the blended spreads, how we see them evolve over time, and that is Page 8. And that gives you the best indication what it means for our fleet.

So Markus, just to make sure I understand, if you look at your Page 8 and I say I'm taking the line that you call calendar 2019, which currently is at €2.5 or €2.4 per megawatt hour, you're telling me that this is the number I can plug against your megawatt and your terawatt hours to understand the profitability of both the Lignite & Nuclear division and the European Power division in 2019 based on market conditions now.

You mean -- you are on Page 7?

I'm on Page 8.

On Page 8, no, Page 8 is relevant for the spread assets. So this is relevant for European Power.

European Power only, okay.

Exactly. And on Page 7, especially the delta is relevant. So you see, we had -- if you see 27 on '18 and 27 on '19, and the position which is in the money around the same, 85 to 90, 80 to 85 terawatt hours, you can assume the profitability in this segment on the margin side will more or less be stable from '18 to '19. And then, we have additional cost measures.

So you have effectively hit the bottom in Power Generation.

In '18, not in '17, but in -- we are clear that we have hit the bottom in '18 because we're going to see a lot of recovery in '19, not only on the -- that we have the full UK capacity market, the additional €300 million in cost savings in the bank, but also when it comes to cash flow much lower utilization of provisions. So we see -- we think -- I mean, we are very confident that we're going to hit the bottom in 2018.

The next question is from Mark Lewis, Barclays.

I have two questions as well but they're related. The first one is, I mean, you talked a lot about carbon in the European context. But we have the German election coming up and we still have the issue of Germany's domestic emissions target for 2020, which is a 40% emissions reduction target, which at the moment seems almost unimaginable really, that Germany can now meet to that target. Nevertheless, in the CDU manifesto of the 4th of July and an interview that Angela Merkel gave to the German media on the 16th of July, she recommitted to that target. Do you think there is a residual risk that there will be more stringent measures put in place after the forthcoming election in an attempt to try and meet the domestic German emissions target? That's the first question. And the related question really, coming back to Mátra, what's the strategic logic for selling that? I mean, given that you are very clear that you've refocused now as an industrial company on those areas and those markets where you feel you have a competitive advantage, surely there aren't any, if any, aren't many, if any, companies out there who have got a clearer idea of running a lignite plant than yourselves. So what's the logic for divesting Mátra?

Yes. I mean, on Mátra, we made very clear what our core markets are. Our core markets are Germany, the German market, so including Switzerland and Austria, the Benelux and the U.K. market. And running a single asset in an environment which is also from a political and regulatory point of view very difficult or even becoming more difficult, is not our core business. And so to be very stringent and consequent, we decided to sell it because there might be better owners around and it also reduces management complexity and it balanced also our portfolio if we exit especially lignite production. So from a strategic point of view, we think it makes perfect sense because it balanced our portfolio better, we exit a non-core market in a political regulatory environment where probably other people are better to operate than we are. On the CO2 side, or the national emission targets, I think that will also be part of the equation. Let's see and wait for the new government, how they want to achieve that, whether they want to achieve it, how they're going to balance it. It is not only a problem of the utility sector. I mean, the public mood is currently also especially focusing on the transportation sector because the only sector where emissions are going consequently down year-on-year is Power Generation and where emissions are going up is the transportation sector, yes. But I mean, you are right. I think, I said it's going to be especially security of supply because there was a question, where could we gain from the policies of the next government? But you are right that also achieving the 2020 national targets will probably be part of the equation. But we are confident, we are already contributing because we are putting the lignite blocks into security reserves, 2 blocks every year from now onwards.

Yes, absolutely, that's understood. But I mean, obviously, to get to the target for 2020, you'd still need quite significant incremental measures. Very clear.

You're right, yes.

Thank you. The next question is from Ingo Becker, Kepler Cheuvreux.

On your Slide 8, I understand correctly that this is an RWE production mix-weighted chart rather than the market average. And drawing from that, do you read anything more into those trend lines than the rising fuel price differential that we've seen or do you detect any supply-demand factors in that as well? And my second question would be on your commercial asset optimization income, I think you said you are assuming a normalized level in H2 and I'm wondering why that is. Did you find the H1 market environment just peculiar, supportive or why don't you think that this might be repeatable?

The first one, Page 8 is not our production mix. It is the market, because the spreads you see depend on market conditions, not on our specific situation. So what this is, is -- I mean, we run a fundamental model, a fundamental supply-and-demand model with baseload assumptions, as I said earlier. Peak load assumptions, we see the current price levels. We then have an assumption how often, which technology is price setting, what the pass-through factors are, and then we can calculate a blended clean fuel spread of the market, which if you use exactly our assumptions on technology mix, pass-through factors and so on, which you can currently buy in the market. That is what you see on Page 8. But of course, that drives the profitability of our fleet. If this is higher, we make more money. The second, on commercial asset optimization, yes, let's put it this way. We are pleased and -- about the performance of commercial asset optimization for the first half and it's a cautious approach to assume that it will fall back on normalized level in H2.

Do you have a reason already into Q3 now to assume that or it's just a matter of being conservative? And if I may ask on the first one, do you find anything else in the rising trends for the fuel spreads than the change in the fuel price differential within, e.g, coal and gas or would you say that basically is it?

No, it is -- that is one factor, but the other one is how do the markets see the tightness. And what we see is -- I mean, we always said the truth is always when it comes to physical delivery because then financial markets are irrelevant. So what you see in the outturn is the physical reality of supply and demand and the financial markets don't play a role. The question is, how early do the financial markets reflect the fundamental reality? And the earlier they reflect it, the earlier we start to close our implicit fuel hedge position. And it's now speculation whether -- why calendar year '20, for example, is already above €3, even 2.5 years prior to delivery. It's just -- I mean, are the current price consolation in the forward markets is [indiscernible]? On the commercial asset optimization result, no, there are no indications in Q3 that make us more cautious. It's a matter of fact that we are more cautious, but especially in Q2 where we say -- where we saw significant price changes, especially in gas and coal, which then is used by the commercial asset optimization team to make an additional euro when they use a new price consolation in order to optimize the portfolio even better. We don't expect this fuel price volatility to continue for the next 6 months.

The next question is from Nick Ashworth, Morgan Stanley.

A couple of questions for me as well, if I may. On European Power Generation, you've talked a lot about the commercial asset optimization and a little bit around the better performance of some of the thermal plant. Can you talk a little bit about the cost side of things as well and how that's moving? I know you have the cost program and targets through to 2018. I just wanted to get a bit of an update from that and what's been delivered so far this year, and if you're happy with the run rate? And then secondly, on the dividend, I guess, looking at it, there's a lot of money coming in at the moment. You obviously have the nuclear tax monies come in. Distributable cash flow was strong in the first half and presumably shouldn't move hugely in the second half, if I read into what you're saying. You're selling Mátra and there's other bits and pieces. You'll have a lot of cash coming in and maybe not as much going out. You've obviously committed to the euro special dividend. But I just wondered on the €0.50 on the underlying ordinary, would there be any upside risk to that this year? Is that something you would perhaps look into for next year?

Thanks, Nick. On European Power, the cost measures, I mean, we promised another €100 million cost saving for European Power until 2019. So far, it is going very well. They are slightly ahead of the linear plan. So the contribution in the first half of the year is very satisfactory. On the dividend side, yes, we are very pleased that we see a very strong distributed cash flow generation in H1, also as you said, maybe not much expected in H2 given that we got the innogy dividend in Q1 and some reversals on the net working capital side. But let me reconfirm what we said before. We will keep our promise that we pay our distributable cash flow in full to shareholders, but we also said that will not happen on an annual basis because we want a smooth development over time and not a flippy floppy dividend payment. And that's why we take a longer time horizon into consideration, let's say maybe 2 to 3 years, and we're going to hit the bottom in 2018 as we discussed earlier. So is there upside risk? I don't know,, but it's too early now. I can just reiterate the promise that we pay our distributable cash flow but it is too early. First, we want to have clarity what the 2017 distributable cash flow will look like at the end of the year and much more clarity about the development in '18 and how big the recovery in '19 is. So the right point in time to reconsider dividend is when we present our full year results 2017.

The next question is from Vincent Gilles, Crédit Suisse.

Me again. Two very precise questions. Number one, you just mentioned it's on working capital, the swing in working capital versus last year in your cash flow statement. So last year, minus €2.6 billion, this year minus €1.8 billion. If you can help us bridge, please. And the second one is on interest income, €125 million in H1. On the cash, I will simplify. I will just take the cash you've got in the balance sheet, €5 billion at the end of -- €4.6 billion at the end of December '16, €10.9 billion at the end of June. It seems a pretty rich, actually, level of return on your cash. Can you just help us understand how you get to a high level like that? And I assume that some of the cash is returned from the government on the 6% interest, but wanted to check with you.

Just to understand whether I got it right. The last question was interest result compared to...

Interest income, the first line of the...

If you look at our reported figures, so not the adjusted one, there in our interest result, the €250 million interest reimbursement we're going to get from the government is reflected in our interest. So there's a positive effect of around €250 million from the reimbursement of the nuclear fuel tax in the interest result, and that is the reason, yes. And the other question was on working capital, but your question refers to the group, right, not with the standalone figures?

To the group, yes.

Yes, that is very easy. I mean, you can bridge -- you have to bridge around €1.7 billion. It is €1.7 billion nuclear fuel tax. It is a better operating performance of a couple of hundred million. And then, of working capital, a positive working capital effect, but that is especially at innogy, and Bernhard has reported that on Friday, because they have seen a very negative development last year in a more normalized but still negative pattern this year. And that is the effect why it changed.

Sorry to interrupt you. The bit I'm interested in is what you mentioned in the middle, just between innogy and the €1.7 billion. You said some improvement in performance. Can you maybe elaborate a bit on what we should expect in the second half of the year?

Now on which figure exactly? I'm not sure whether I...

On the working capital.

Are we now talking about the standalone figures or the group figures?

No, I'm still with the group. Do it on standalone if you feel more comfortable. It's just to understand the swings we can expect. I can deal with innogy separately, is no problem.

So then we only -- I mean, we should expect a more normalized level where we said on a year-to-year basis in a very usual year, we don't expect any significant working capital effect which are -- so more or less neutral. But in addition to that, we said we're going to phase out some optimization measures given the very healthy liquidity situation we are in, and given the interest environment, these are currently not economically favorable anymore. So overall, we should expect a slightly negative effect on the working capital side for the full year. And currently, we stand at plus 0.2.

Just if I can keep you for one second on the tax rates, sorry to ask a third question. But just to -- should we -- what should we expect for the tax rate to the full year on a group and on a standalone basis?

Group is very difficult to predict, almost impossible and also, I think, economically not very relevant. Because, as you know, for innogy, it depends on where they make their money, which jurisdictions and they saw a very high tax rate in the first half of the year because most of the money was made in Germany. So -- and then if you combine that with our result where we actually don't pay any taxes, the combined tax rate is meaningless and very difficult to predict. So let me focus on the RWE standalone side. There, we actually don't expect additional taxes for H2.

The next question is from Deepa Venkateswaran, Bernstein.

I have 2 related questions on the commercial optimization. So you mentioned earlier that you see an opportunity of £2 to £3 per megawatt hour from optimization. How does that square with the middle double-digit number that you've revealed? Because FIC for the half year, you've basically had around 47 terawatt hours in your European Power Generation. So I was just wondering if you can help square that. And the second question is your Chart 8. Would you be able to share the pass-through factors using that chart? And if not, at least can you say whether these factors are constant for every single curve or are they changing? And if so, what's the direction on the change?

The first one is, they are definitely changing from year-to-year because we have different assumptions for the supply-and-demand balance and different prices for the fuels. But unfortunately, I will not share any detail because that would put you maybe in a position to find out what our fundamental assumptions are. And I mean, this model is very important for us and is definitely a company secret. So we will not give you details on the exact pass-through factors we assume because you could then calculate what our assumption on demand and supply is. On the CAO side, it's €2 to €3, euros, not pounds, per megawatt hour. And I mean, on Page 7, you can see how much, what is the production on a yearly basis, around 100 terawatt hour outright position, around 50 to 70 terawatt hour spread position. So you can calculate the overall amount CAO makes and we said the additional amount they made in the first half of 2017 is a medium double-digit million number. So it's around 10% more than we usually assume on a full year basis. So already 20% more on a half year basis.

I believe there are no further questions.

Good. Well then, thank you, everybody, for joining the call. If there are any questions, as you know, we are available to take your calls 24/7. Thanks a lot.