Deutsche Telekom AG

Thomas Dannenfeldt - CFO Hannes Wittig - Head, IR

Yin Tang - UBS Investment Bank Dhananjay Mirchandani - Sanford C. Bernstein Akhil Dattani - JPMorgan Chase & Co.

Ulrich Rathe - Jefferies LLC Stephane Beyazian - Raymond James Euro Equities Mandeep Singh - Redburn Robert Grindle - Deutsche Bank AG Frederic Boulan - Bank of America Merrill Lynch Ottavio Adorisio - Societe Generale Matthijs Leijenhorst - Kepler Cheuvreux Russell Waller - New Street Research Andrew Lee - Goldman Sachs Group Mathieu Robilliard - Barclays PLC Samuel McHugh - Exane BNP Paribas Justin Funnell - Crédit Suisse AG Stefan Borscheid - Landesbank Baden-Wurttemberg

Yes, good afternoon, everyone and welcome to our First Half and Second Quarter 2017 Conference Call. With me today is our CFO, Thomas Dannenfeldt; Tim Höttges, Our CFO -- CEO is excused today because he's enjoying his long-planned and well-deserved holidays.

He will -- Thomas will first go through a few highlights and then he will discuss the quarter's financials. And after this, we will have time for Q&A.

Before I hand over to Thomas, please pay attention to our usual disclaimer which you will find in the presentation. I hand over to Thomas.

Thank you.

Yes. Thank you, Hannes and a very warm welcome also from my side.

I think there are only 2 things I'm not going to do today; one is to participate on the speculation about the U.S.; and the second one is to tell you where Tim is doing his holidays. So -- but seriously, I'm very pleased to say that at the half-year stage, we're well on track for our 2017 target.

We've seen good commercial and also good financial momentum in all our markets. We've upgraded our EBITDA guidance today due to the T-Mobile U.S., but we're also seeing a bit of growth on this side of the Atlantic which is important to us as well.

So let's have a look into the presentation. On Page 4, you can see that our peer-leading momentum continues.

Our customer growth is strong across all geographies. Our revenues are up 5.9% year-to-date and our adjusted EBITDA, 8.2%.

In Germany, we're now -- have over 8 million fiber homes, having added 1.4 million this year alone. In the U.S., we delivered another strong quarter and what is particularly pleasing, we also saw strong customer growth in our European subsidiaries.

We keep investing at record levels. After investing €11 billion in 2016, our year-to-date investments are up 13.5%.

But because of our high EBITDA growth, our free cash flow is still up 18% year-on-year. And despite our USD 8 billion investment in over 30 megahertz of U.S.

low band spectrum, we managed to retain -- to remain within our net debt to EBITDA comfort zone. So we're well on track for another year of major investment to the future, plus double-digit free cash flow growth.

Today, we raised our EBITDA guidance for 2017 and we keep delivering well against the targets of our 2015 Capital Markets Day. Slide 5 shows some examples for the strong momentum we're seeing with our customers.

In the last 12 months, we added 1.6 million converged customer, 2.6 million German fiber customers and 6.5 million U.S. customers.

In the cloud, we grew 11% year-to-date after an acceleration in the second quarter. As you can see on Page 6, we maintained good momentum with our innovative product.

For instance, we added 97,000 customers to our smart home platform in the last 12 months. In the quarter, we launched our new router which integrates our smart home platform now.

Our new StreamOn tariff in Germany, launched in April, has proved very popular and already has close to 300,000 subscribers. This week, we revealed our new telecom sports package, based on exclusive arrangements and agreements with Sky.

For €9.99, a month, our Entertain TV customers can now watch live conferences of the Bundesliga, the Champions League and the German Handball League. This is in addition to our own content such as the national ice hockey, basketball and the woman's footballs league.

Also Germany, our new entry-level TV product, StartTV, is off to a good start and has contributed to over 69,000 TV net additions this quarter. We've also launched HD voice+ to provide our German customers with a superior voice quality.

In the U.S., we gained demonstrated technology leadership and launched LTEU and are testing LAA. Both technologies are taking advantage of the unutilized spectrum in the very wide 5 gigahertz band.

This band is, today, primarily used for Wi-Fi. We believe this is an excellent opportunity to complement our strong, low and mid-band spectrum positions towards a superior customer experience.

Talking about the U.S., you will remember our statement in our first quarter conference call that it's possible, if not likely, that there will be various discussions regarding various potential strategic combinations among industry participants, including ourselves. We said at the Q1 stage, it is of course far from clear that any such discussions would lead to anything.

And in any case, our 1 priority is to leverage the fantastic opportunity that we have at hand and that we just further strengthened with our great success in the low-band auction. You know our policy is that we do not comment on M&A speculation and we and I will stick to our line that it would be wrong -- and that we will not give running commentary here.

By the way, I believe there are anyway enough people speculating around. Slide 7 compares our year-to-date performance with our guidance.

It is fair to say that we're comfortably ahead of the group levels guidance we issued at the 2015 Capital Markets Day. For 2017, we now incorporate the recent EBITDA guidance increased by T-Mobile U.S.

The U.S. has increased the EBITDA outlook by USD 100 million and the outlook for the remainder of the business is unchanged so that we can raise our group EBITDA guidance to €22.3 billion, accordingly.

Just a few words on the impact to the free cash flow. After T-Mobile, our guidance increased €50 million due to an increased outlook for handset lease revenues.

This does not have an immediate cash impact. Also T-Mobile now expect that CapEx will be at the upper end of their guidance range of $4.8 billion to $5.1 billion due to early 600 megahertz deployment which makes a lot of sense.

We therefore keep our free cash flow outlook for 2017 unchanged at €5.5 billion, up from €4.9 billion last year. So let me now discuss our second quarter in a greater detail, starting on Slide 9.

Adjusted EBITDA growth accelerated to 8.9% this quarter and to 8.2% at the half year stage. Our adjusted net income grew 40% this quarter.

Free cash flow was slightly down year-on-year this quarter, but this is just phasing. For instance, we incurred roughly €0.2 billion of core premium in the context of a recent bond refinancing in the U.S.

At the half year stage, our free cash flow is up €4.4 billion or 18%, so we're well on track for this year's €5.5 billion free cash flow guidance. Moving on to Slide 10, in Germany, our sales were again slightly up this quarter, but total service revenues were slightly down due to various regulatory headwinds.

Our EBITDA grew 1% this quarter at the half year stage. While this looks a bit slower than what seems needed for the year, this is no concern for us and no cause for concern.

For instance, in the second half, we will benefit of the -- we will benefit from the merger of our sales and technical field service that we announced last years and associated savings. So we see ourselves very well on track for our guided €8.4 billion EBITDA for the year as a whole.

On Slide 11, you can see that our reported total service revenues are slightly down. But when you adjust for roaming and for both fixed and mobile termination rate cuts, our total service revenue would have grown 0.3% this quarter.

Looking at mobile, our service revenues grew 0.8% this quarter. The sequential improvement mainly reflects growing data usage, while 0.5 percentage points of the 1.6 percentage point's quarterly improvement are due to a lower regulatory drag than in Q1.

We also saw a nice growth in M2M where we added almost €1 million since the last 12 months. Without regulation, our German mobile service revenue would have grown 2.5 percentage -- 2.5% this quarter.

Both our B2C and our B2B mobile sales revenues improved sequentially. And as always, small quarterly trends variations should not be overrated.

But we see ourselves very well on track for our Capital Market guidance of percentage -- of 1% CAGR before roaming. Our fixed line service revenue momentum worsened slightly this quarter, mainly due to a weaker broadband revenue growth and a difficult comp in wholesale.

There's more detail in the appendix. Despite various competitive promotions, our commercial performance has been very steady.

We lost 186,000 contract customers this quarter, but our own brand customer base grew by a strong 228,000, slightly stronger than in the more recent quarters. Our new 3-month product had a positive impact.

As in the previous quarters, almost all of the volatility here is due to low-end cards, such as Lebara and not service provider business. So it has basically, no relevant revenue impact here.

As you know, convergence is our strategic focus and we'll continue to make good progress here as well. This quarter, we further increased the percentage of homes with the MagentaEINS contract to 17%.

And 38% of all Magenta-branded mobile contracts are part of a convergent relationship. Our up-selling remains very successful; homes signing up to MagentaEINS spent an average €8.60 per month more with us than before.

Benefiting from our various innovative propositions, our mobile data usage growth remained strong and contributed to our improved mobile service revenue performance. This remains a focus for us.

You may have seen that we have raised the data allowance in our S&M tariff with effect from next week on buy 1 gigabyte. Maybe 2 quick comments on this.

First, this is a time-limited promotion for the summer months and second, while this looks like a more for the same, it's actually a more for more move, because we have raised 1-off fees and taken out certain promotions. Moving on to next -- to the fixed line on Slide 14.

Demand for our fiber products remains very strong with over 600,000 additions this quarter. Again, it's like 7 -- I think it 7,000 a day which is an amazing run rate.

While our resellers keep doing well, most of the new customers were again on our retail platform. We continue to think that this strong customer response clearly demonstrates the benefit of our strategy to bring our superdense fiber network to as many homes as quickly as possible.

TV net adds were strong again on 69,000. This also benefited from the launch of our new StartTV.

We added only 46,000 broadband customers this quarter. Our growth was again affected by our hard IP migration which for legal reasons, involves canceling a customer line, still unbelievable.

Adjusted for this headwind, we will have added well over 70,000 broadband customers this quarter, consistent with recent performance. The sequential slowdown compared to the first quarter reflects some seasonality and slightly greater hard migration headwinds.

We continue to target similar broadband customer growth in this year as in 2016. But clearly, whether this can be achieved will depend on how successful we're with various plant mitigation matters.

Line losses were also hurt by the hard migration. In addition in May, we put through a small line rental hike for single-play customers.

The resulting churn was a bit better than what we had expected, but largely explained the increase in line losses the first and the second quarter. Our broadband revenues grew 0.8% this quarter, after 1.4% in Q1.

The slowdown in the run rate reflects the still progressive impact of the promotions launched last summer, plus the incremental headwind from the hard migration which is worse than we had originally expected. While we expect trends to improve in the coming quarters, it is fair to say that from today's perspective, the target of a 2% CAGR in broadband revenues which we promised at the Capital Markets Day, looks quite challenging.

On Slide 16, you can see that we now pass 67% of the German households with our vectoring network. We're on track to launch super vectoring in mid of '18.

This will increase the speed for most homes in our footprint to 250 megabits per second, at a fraction of the cost of the initial FTTC deployment. 61% of our excess lines are already on IP, up from 47% one year ago.

This accelerating run rate takes us towards our target of migrating our B2C access lines to all-IP by the end of 2018. Now moving on to our usual 2 slides on T-Mobile U.S.

who have already presented very strong numbers 2 weeks ago. Our momentum, as you know, remains very strong.

We won 0.8 million branded postpaid phone customers. We also added 0.1 million prepaid customers, with a sequential improved ARPU of $38.70.

Service revenues grew 8.5%, while IFRS EBITDA grew 18%. And as mentioned, T-mobile raised its EBITDA outlook for 2017 by $100 million.

On the next slide, we'll show some selected performance metrics for our U.S. subsidiary.

Our branded postpaid phone churn improved further and is now at a record low of 1.1%. Our bad debt expense ratio dropped significantly, while our cost of service remained at 20% of service revenues.

As you can see, based on user generated data but strongly underscored by yesterday's open signal report, our network has a clear and growing lead on the competition. Since their launch of unlimited, both Verizon and AT&T have seen their network performance deteriorate, while T-Mobile's has further improved.

A key focus for T-Mobile has been our commercial expansion towards 17,000 brand doors by year-end. This is an increase of 3,000 of which just over 2,000 have been delivered year-to-date.

Another focus has been small sales, where T-Mobile is planning 25,000 next 18 to 24 months, of which all will be fiber backhauled. To sum it up, another great quarter for T-mobile U.S.

and lots of potential going forward. Back in Europe, it was a strong quarter for customer growth in our Europe segment.

Continuing the recent acceleration, the segment added 372,000 new contract customers and 175,000 new converged homes. In Poland, where we had been losing share, we're now turning the tables and saw improving porting ratios and another positive quarter of contract net additions.

Meanwhile, our European TV and broadband momentum remains strong and steady as well. Our revenues in Europe were up 2.4% this quarter.

Mobile service revenue grew in most markets, accelerating to 2% year-on-year across the European footprint. Reported EBITDA was down 2.2% but, adjusted for our recent reorganization in FX, EBITDA would have been stable, consistent with a trend we continue to expect for the year as a whole.

The next chart shows that we now have migrated 64% of our homes to IP. Our LTE coverage now stands at an impressive 91%, up 7% year-to-date.

And our fiber coverage has reached 27%. Moving on to systems solutions, on Slide 22.

Here, we saw small revenue decline as we continue to execute our plan to move away from classical IT towards network-centric services. EBITDA was up strongly year-on-year, but this is mainly reflecting normal quarterly volatility and, as guided, we continue to expect a broadly stable EBITDA performance for the year as a whole.

The next slide we show our new segment, Group Development which mainly consists of T-Mobile Netherlands and our German tower business. Importantly, from the beginning of this quarter, STRATO was no longer included from the 1st of April, so we're showing the numbers on this slide without STRATO.

We had another good quarter in the Netherlands, with 61,000 contract net adds. This is now the eighth consecutive quarter of contract customer growth.

Dutch service revenue were up year-on-year. EBITDA also benefited from efficiency measures and from recent change in Dutch consumer protection laws as of the last quarter.

Our tower business reported a slight year-on-year EBITDA decline and was mainly due to a transition cost following the carve out of our German segment. Overall, the segment EBITDA was almost stable once you exclude STRATO and is well on track for our full year guidance.

As already mentioned, our second quarter free cash flow was slightly down year-on-year, due to phasing and team U.S. refinancing costs.

But at the half-year stage, we're up 80% and we continue to target the €5.5 billion for 2017 as a whole. Growth in our adjusted net income was driven by a higher adjusted EBITDA and a better financial result, but we're benefiting from the improved interest environment.

Higher taxes and a higher share of the minorities, driven by the strong performance in the U.S, partially offset this positive impact. Our net debt benefited from the growth and free cash flow, the STRATO disposal and also obviously, FX, but was negatively impacted by the $8 billion spectrum payment and the TV.

Slide 25 shows our financial metrics. And as mentioned, despite the U.S.

spectrum auction, our net debt remained within our comfort zone of 2x to 2.5x adjusted EBITDA. And I can tell you we're particularly pretty pleased with that as well.

As always, my final slide summary -- summarizes the strategy represented to you at our last Capital Markets Day. We continue to execute well against our targets.

However, I think the time has come to give you our perspective for the years past. Our current guidance will rise and therefore, Hannes and myself and also, in Tim's absence, Tim, we would like to invite you to our next Capital Markets Day which we'll have scheduled for the 24th and the 25th of May.

And this having said, I think we're going to go to the Q&A.

Yes and the good news is that Tim will be there at the Capital Markets Day. So thank you, very much, Thomas.

We can start with the Q&A. [Operator Instructions].

So with that, we go to the Q&A. I think I can see already a few questions here.

So the first one would be from Paulo at UBS, please.

I've got two questions. The first one is just on German mobile.

There's a notable improvement in terms of mobile service revenues in Q2. Part of it was lower EU roaming drag, but there was also notable underlying improvement.

So how should we think about the quantum of roaming drag through the rest of the year? But also were there any 1-off factors boosting underlying service revenue growth in Q2?

Or is a 2.5% underlying mobile service revenue growth sustainable over the coming quarters? That was first question.

Second question was raised about all-IP migration economics. So can you give us a sense in terms of the quantum of the OpEx savings that you can get per line when you make -- migrate them from your current network over to all-IP?

Yes. Thanks for the question.

First of all, on the mobile service revenue, first quarter headwind on the roaming was around 21% -- €21 million. Second quarter is around €12 million.

So we have a little bit of support this quarter on -- due to that fact that, obviously, the headwind will increase in Q3 again, so that's 1. So that was quite favorable for this quarter.

And as I said, we'll stick to our 1% CAGR ex-roaming here. Obviously, I said on the broadband side, the 2% are quite challenging.

The 1% CAGR on the mobile side, that's where we feel comfortable and might be a little bit higher. But that's basically where we're.

The full year impact on roaming, just to add that, will be 17 -- €70 million to €80 million. So again first quarter €21 million, second €12 million and then the remainder of the year you can calculate.

That's on the German mobile. And the second question was IP migration OpEx saving 1 line, migrating to all-IP.

I think we haven't disclosed that for the -- especially for the German environment. Especially, because basically the savings you get if you switch off platforms.

So as long as you migrate the customers, there's no savings, there is extra costs and that's what we're still facing right now. Once you have all the customers off the platform, we switch off, you will see then the effects of -- it's hard to do that per line.

It's much more related to a platform perspective. And remember, what we've given you in the Capital Markets Day was that the extra cost we will have will peak in '16 due to some delays we've seen in the B2B area.

We said '17 will be like the same kind of vicinity. But from here on, we're just looking forward, the profile should improve and that's what you should keep in mind here.

And finally -- that's a good, thank you, Hannes, for the remark. And finally, the complete effect that, meaning Germany and Europe, after a complete transformation will be €1.2 billion net OpEx reduction.

We will see that in the early 20s then.

So the next question, thank you, Thomas, is from Dhananjay at Bernstein.

My question is a built up on Polo's question related to the all-IP migration in Germany which is a crucial part just given the orders of magnitude of the savings to your structural OpeEx reduction story. So given that 47.7% of retail lines are still on legacy technology and that you've been averaging about 3.5% in IP adoption per quarter of your retail line base, you still have arithmetically three years to go versus what you mentioned in terms of your target of completing the transition by 2018.

I'd just love to understand how to reconcile the difference because it suggests that you're planning for some sort of a hard lines migration. That's the first half of the question.

The second half is, how concerned should be we about the implications of such a hard migration on the flow through of your aspired savings on a run-rate basis to EBITDA.

First of all, let me again clarify the numbers. Currently, we have 61% of the lines migrated.

So it's like 39% to go. We're, for the mass market, absolutely comfortable that by the end of '18, we will have that done and completed.

And as we said, I think several quarters ago already in B2B, we might slip a little bit, talking about end of '18, because it's much tougher to migrate those bigger complex -- more complex customers. But in terms of the mass market, I think we're not worried about that, getting it completed by end of '18.

I think the implications right now are that, due to that specific German law, we're asked to really cancel the line of a customer first. Normal behavior would be, you inform the customer about a change in the line and you give them an opt out, but -- and it would be much more customer friendly and what you normally think some players should do with his customers.

But we're required really to cancel the line. So 1 implication is -- and that's what we're seeing right now -- is that driving that hard migration which is necessary to get the people off the platform finally also includes some elevated churn on the broadband side.

And that's -- the number I've given its last quarter was high 20s -- 20,000s, what we've seen there. And obviously, there's a big focus on our side to find ways to mitigate that.

But still, be compliant with that, from my point of view, strange legal situation we're looking at. I'm not sure whether that answers your question complete but I think so...

Okay, so Dhananjay, does this answer your question?

Thank you.

Okay, that didn't work but I hope you can e-mail me and we can pick up if you have a further question. So I move on to Akhil at JP Morgan, please.

Two questions, please, if I may. Firstly, on the StreamOn launch in a Germany.

Tom, you mentioned that you've seen good initial customer reaction, I just wondered if you could maybe help us understand exactly what you're seeing? I guess either in terms of customer mix, upsell, whatever the kind of key metrics you're seeing or what flags in terms of what you're seeing at the moment albeit, appreciating its early days.

And the data traffic growth in Germany in your presentation slides seems to have sequentially slowed. I just wondered why that was; it would seem a little bit counterintuitive with that launch but maybe there's something else going on with MVNO's or wholesale or something else.

So that's my first question. And then the second question, I guess it's just really a question around currency.

Also, we have seen quite a big move year-to-date on the U.S. dollar-euro rate.

Just keen to understand, when thinking about earnings and free cash flow, how we think about the drop through in terms of debt exposure, hedging, things like that? Just so we can kind of quantify the impact on numbers.

Okay, thanks, Akhil, for the questions. First of all, on StreamOn, as mentioned, we're now around 300K customers on that one and I think it's important to understand that we put the -- let's say the barrier to entry relatively high.

It's for customers in the -- from the tariffs scheme from M upwards. So we haven't opened it to the complete base so to say.

But clearly from that level upwards. Basically, we see three elements, new customers upsell into a higher tariff plan and people just using net optionality within their tariff scheme they're already in and having a better experience.

We're comfortable and fine with the mix; the new customers, that's around the vicinity of 20%, roughly. So it's not the driver of the whole thing, but it's also contributing well to it.

We see a share of upselling in there. So a pretty good mix of making customers use more upsell and also attracting some new customers.

Of course, the data traffic growth in Germany has slowed down. I don't -- I'm not sure -- 61 -- maybe Hannes you can help there.

Yes, just on...

61 and before it was around the same vicinity.

Yes. Just when you look at Chart 13, you can see that, while it's true that the year-on-year rate has decelerated, the increment of the second quarter relative to first quarter is actually greater than the increment of the first quarter versus the fourth quarter.

So we do see some traction here from the StreamOn. But of course, you can also do the math.

It's fairly early days for the StreamOn to have a full impact, because it's ramping up at this point. So in terms of your other question on the U.S.

dollar, how that feeds through to the bottom line on net debt, I think of course, we have some real hedges in the -- of the debt. We have $30 billion roughly of the debt, it's in US dollars.

So you can make the math there, so it's had some natural hedge in our net profit figures.

Yes. And then the guidance is based on constant currency and the FX effect, I think it is a -- on a net debt there's a -- on the debt, there's a detailed information in the backup, yes.

Okay, so let's move on to Ulrich at Jefferies, please. Ulrich, can we have your questions?

My first one is on the mobile Intec -- the brand of Intec was quite strong. Was wondering whether you're willing to talk about the Congstar share versus the Magenta share?

And I think in the past, it was sort of roughly give-and-take 2/3 Congstar. Is that still the case?

Or has it ticked up and drove the uptick in the second quarter? My second question is this, on the all-IP migration, you hinted towards some mitigation measures that you're planning, I think, in the second half.

Could you sort of talk about that a bit more in terms of what you're trying to do and whether that would sort of even touch price points? Or what sort of measures you're short of trying to put in place to mitigate the issue?

Okay, first of all, on -- thanks Ulrich, for the question. For -- on the Congstar share versus Magenta share, as you know, we don't disclose that but it is constant more or less around the last quarter.

So there's no big swing in the one or the other direction on Congstar versus Magenta. So no big change here.

On the all-IP migration, mitigation and discussions and then thoughts about mitigation are not around price points. That's not what the customers' worries are.

Basically what we're doing is, we're approaching customers who are happy with our product and they don't want to have a change. And then we approach them and we cancel the line.

And as I said, unfortunately, what we can't do is tell the customer "You don't worry, we just switch you to IP and you don't have to do anything, we'll take care." What we basically do is we cancel the contract and then some days later, we can come up potentially with an offer and say, here's a new offer.

And it's much more about the way we approach the customer, make him understand that we will help him through the process and make sure that, first of all, we don't give too much negative messages and then come back and try to fix it which as I said is very much regulated and required by law to go that 2-step approach. So the whole mitigation is much more about more -- let tell you -- say it like this, more intensive care about what's going on in that process rather than pricing.

Okay, thank you, Thomas. The next question is from Stephane Beyazian at Raymond James, please.

One question regarding the cost in Germany. I mean, you're clearly delivering on the return to the EBITDA growth in Germany.

So I'd like to understand how sustainable this is despite -- with your headwind coming such as roaming? So are you able to elaborate a little bit on the outlook for some of the moving parts in the cost base, such as staff cost, marketing and acquisition cost, network cost for the remaining of the year and perhaps beyond 2017?

Yes. As you know, in the first half, we grew 1% and the guidance is grown from -- growing to from 8.2% to 8.4% which is just kind of 3% in the second half of the quarter.

In the first half, we mainly benefited from the significant headcount reduction in 2016 we had which was impacted by the front loading of civil servants early retirement program. In the second half, we expect EBITDA growth to accelerate.

Why is that? Several reasons, there are a lot of further IT on process optimization, automation activities going on.

You can also talk about that, as part of our digitization, so those kind of effects. Efficiency is also related to the merger of our customer service and field service organization we announced last year which obviously we have done to get better output with a less input, meaning less cost and better customer quality.

And that kind of initiative is also, in terms of phasing, will show the delivery and the result in the second half of the year. So that's why we're comfortable for the guidance of 8.4%.

And I think just to make clear the roaming as such, Thomas has already talked about the roaming impact has already been, let's say, 40% of the full year impact in the first half. And the full impact, we said is around €70 million.

So yes, it will be a bigger drag in Q3, clearly for obvious reasons. But it's not, let's say, a significant item when you look at the overall magnitudes of the growth that we -- in EBITDA, that we try to achieve for the year as a whole.

Let me move on to...

Wait a second, before we move on, Hannes, I've forgotten 1 element of the -- or 1 answer. Part of the answer was also or part of the question was, what to expect beyond 2017, so in 2018.

And I think it's fair to say that we believe there is more to come in terms of cost reduction, obviously from 2018 onwards. We will also see some more support on the EBITDA by revenue because the regulatory drag will more and more go away.

But for sure, I think there is a big opportunity for us to deliver also even more on the cost reduction side. So you should expect some more results here as well.

Yes. And now I move on to Mandeep at Redburn.

Mandeep, can we have your questions please.

Hannes, it was just a follow-up really on -- a little bit related to the previous question. Can we get a little flavor for the sort of scope and quantum of restructuring charges?

They were bit lower now. I understand they're going to step up next year but sort of looking out into the medium term, what's the right sort of number people should be thinking about into the medium term?

And when -- and how many years could we expect like a step change down? Can you just give us some idea of how we should think about this over the next few years, please?

First of all, the long term picture is that I think we will see -- just by the age structure, we will see in the first half of the next decade, we'll see a lot of support by people retiring just by age without additional money we need to spend for restructuring. Within this decade, we will still see restructuring costs.

But as we guided in the Capital Markets Day, you won't see that elevated level you have seen last year, for instance, anymore and you shouldn't expect any peak next year as well. As guided, there will be that kind of plausibility level.

But I guess there will be a kind of stable to slightly declining profile looking forward.

Yes. And this year, we'll be lower than last year for the year as a whole as well.

Yes.

Yes. Because last year if you recall, we had the civil service retirement -- retirement program that we have referred to earlier and that was a bit of a lump that's favorable for us.

But it's also because we worked through it last year, but it's impacting year-on-year performance. So let me move on to Robert at Deutsche Bank.

Robert, can we have your question please?

The first is on the cloud revenue growth which seemed to have quadrupled in Q2 versus Q1. And I know last year was a bit soft on that front, but has anything changed in the quarter?

Was it just a big contract or was there a favorable comp? Some color would be great.

And then just going back to the strong mobile growth in Germany in the quarter, was there any benefit from customers recontracting with handsets who had been SIM-only previously. How has the SIM-only handset mix changed in the quarter?

First, to the latter question, no, there was no significant change in the mix of SIM-only to SIM-handset-based prolongations, no big change in the mix. Normally, those mix only changes, if you have a launch of, let's say, like a new iPhone or Galaxy sometimes.

But as I said, no big change in that mix last quarter. And cloud revenue, it's quite simple, the cloud revenues are a mix of public cloud and private cloud revenues.

And the public cloud revenues are that's a -- it's a platform business. It's a quite -- there's not a lot of volatility in there quarter-by quarter, whereas the private cloud is also depending on larger customers you bring in or changing something.

So you have a kind of normal volatility on the private cloud revenues and that's what happened here. There's nothing special.

It's just the nature of the private cloud business that there's volatility on a quarterly basis.

Great. So with that, let me move on to Frederic, Bank of America.

Sorry, Thomas, I'm going to try a question on the U.S. Hopefully, you can answer on that.

So we've seeing a lot of difference permutations being floated around. I mean, is there any scenario that you think would substantially change your ability to drive rate until integration synergies with Sprint?

And secondly, if you can talk a bit more about the German mobile business. We've seen qualitative moves from Vodafone or O2 in the quarter, including much bigger data bundles.

How do you think it's going to shape up? Do you think your current price grade is competitive?

Or you think StreamOn, et cetera, is addressing that? Or eventually we're just moving to more and more high bundles?

So if you could comment a little bit on that phenomenon.

Thanks, Fred. On the U.S., as you know, I'm not going to comment on Sprint, specifically.

I think what's clear is and you know that there is a fantastic standalone position we have in right now and all the ingredients for our future success on standalone perspective are available and I think John is absolutely spot on to focus the organization on the customer in the market and keep going. That's absolutely the right focus.

On the other hand, we all know there is upside by intramarket consolidation. That holds true for Europe conceptually and for the U.S.

and there's also a business case for other combinations conceptually. No news here.

And as we always said, it depends on the specific potential situation. We're confronted, well and we will make up our mind if there's something to think about and then come to conclusion and then talk.

That's the order we doing it. On the German mobile, first of all, bigger bundles, that's what we like.

Talking about network experience and differentiating by network experience. It basically means that customer experience the value of a fantastic network versus an okay-ish network.

And once you see more and more video usage, the difference between an okay-ish and a very, very good network becomes very obvious. You can't cheat your eyes.

They see it. And so for us, driving bigger data bundles is good strategically and it's exactly along the line of our positioning we have.

And is our price grid competitive? I think if you look at our numbers we have disclosed on the net adds on the branded ones, you see, it is very good results.

So there's nothing to worry about in terms of commercial momentum we're having. But clearly, data allowance will go up for sure and we will make sure that we follow the more-for-more logic as we do right now.

And make sure that the willingness of customers to pay is used for better quality. StreamOn is an element of that.

It is -- as I said a minute ago, it is partly addressing new customers to direct new customers. It's partly attracting customers to choose richer bundles and move them up.

And yet, there's also part in there for already high-value customers to enjoy more network. But it's nothing to worry about from our point of view, something we're going to drive actively and kind of move towards bigger bundles.

Very good. So let me take one question from -- that's been sent by e-mail.

So it's from John Dann. I'll take two of his questions.

So one is on the dividend. So the question is, is the dividend linked to all controlled cash at the group level?

Or therefore, there's a full cash dividend need in the U.S -- T-Mobile U.S. to distribute dividend?

Or otherwise return cash to Germany? And maybe I'll start with that one and I'll take the second one.

So that's easy. If you heard Thomas just now, the answer is no.

And clearly behind that is a certain outlook also for the business excluding the U.S. and where we're confident in our free cash flow generation as well.

And secondly, there is a question again from John as well on the German broadband market and he wants to know if -- as the anniversary of the pricing -- the test the best promotion approaches, whether we already have evidence how many customers will upgrade or not. I think, unfortunately, when it comes to test the best, we don't have much experience, of course, because it was launched August last year.

And so it will only begin to roll over from August from now onwards and we will then collect these expenses. But we have some experience related to the previous promotions we had 24 months and €10 rebate for awhile for our broadband offers.

And those began to rollover as of September 2016. And they are, let's say, a large majority of people -- I'd say, somewhere around 3 quarters or more have decided to -- not to negotiate and stick with their -- with the deal and the bargain.

But of course, we can't see that currently in the numbers very well because of the promotional drags that we have in the broadband revenue line. Hopefully, as most promotions wash out, we will see an improvement here in the coming quarters.

Yes and then just my speech said that we're expecting in the second half of the quarter an improvement versus the Q2 numbers. So obviously, that will be a contributor of that.

But as Hannes said, we'll come back to you next quarter with greater details about what really happened on it.

Good. Let's move on to Ottavio at SocGen.

Ottavio, can we have your question, please.

A couple of questions on the other segments. So the first one is on the European one.

This quarter, you recorded very good trends compared with previous quarter. And the 2 main drivers toward the improvements you recorded in Poland and Hungary.

And Poland was very impressive. You went from minus 10% to plus 10% on EBITDA.

But I see that the SEC have fallen by 70% to 80%. So my question is, how sustainable is that improvement?

And are you going to invest back in the market at some stage? And the second one is on the tower business.

The towers are normally business that are relatively easy to predict and relatively stable. But it looks that the EBITDA on your segment has been under pressure over the last couple of quarters.

I believe that you -- it's down to some integration cost. If you can talk more about the each of these costs and if they are going to continue over the next few quarters.

I'll start with the towers question. You're right, it's integration costs and it will continue for 1 or 2 more quarters, but that's it.

So it's not like a long term thing as integration cost shouldn't be. That's the second question.

The first question, on -- I'm not sure whether I got it right, on Poland and Hungry. Maybe you can help on it, Hannes?

Yes. I think on Poland, I think we have -- the second quarter is a nonoperational situation we have -- had last year.

We had a very strong first quarter and a soft second quarter. So that drives the year-over-year trend here.

And the reason for this is, as we highlighted the time, is that we had particularly strong visitor revenues in Q1 and that impacted phasing between Q1 and Q2. So if you look at the -- for instance, the EBITDA trend on the -- for Poland for the half year as a whole, it's a much less spectacular than the Q2 trends.

So it's 2% EBITDA growth for the half year. That's, by the way, on euro terms, so in local currency, it's slightly down.

So Poland, I think the real point for us is that we want to accelerate our -- get back to a fair share of market growth. And we have historically underperformed here and I think it's important for us now to come back, especially, in the B2C market.

And we're seeing good traction here. Thomas has mentioned have seen good porting ratios versus certain competitors who have historically taken share.

And I think this is encouraging and clearly a priority for us. I mean, you can see this as across the European footprint.

But overall, I think when you look at the quarterly performance, half-year performance across the European footprint, there is no particular items to highlight. It's a wash and for the full year.

We believe the EBITDA will be stable, as we have said, adjusted for the accounting change. Okay, so let me maybe move on to Matthijs Leijenhorst at Kepler.

Yes, much of my questions have been answered. Are you willing to share any views on the possible merger between Drillisch and United Internet -- short term and mid term impact?

Yes, for sure. I think, first of all, it's important always to be aware that talking about Drillisch and United Internet, you talk about 2/3 of the market.

There's 1/3 which is B2B market and, by the way, parts of our very good performance in the mobile service revenues is due to very good performance and disciplined actions in the B2B market. And then you have 2/3 on the consumer market.

And that's where we're looking at, talking about Drillisch and United. I believe as they are strong in those segments where we're either heavily underindexing or having just our fair share with Congstar, there is not so much of a direct impact.

It's obviously potentially indirect impact if price pressure is high in those segments. But there's nothing, no changes, basically, to what we've seen in the last quarters on that kind of price pressure.

That's 1. 2 is, I think for sure, the German market some years ago was overcrowded.

It was overcrowded by 70, 80 MVNOs. It was overcrowded by 4 MNOs.

It's good to see that consolidation takes place not only on the MVNO level but also on the MNOs, the small and the bigger ones, that makes a lot of sense. And so for us, that's basically the view.

So another question from the Internet -- off of the e-mail, rather, is from Steve Malcom at Arete and he asked, whether we should expect any material change in strategy with the new CEO for Germany? So over to Thomas.

I don't think so. I think Dirk, who's well known here and we've worked for very long time closely together.

He -- I think he understood that the story is -- and he shares that the story is quite simple. And it is basically network differentiation, service experience differentiation.

And that's kind of the play we're playing. And that execution is the most important element of bringing that to a success.

I think he has gathered a lot of experience last 2 years on Rogers in terms of how it feels to work on a cable basis. To understand that better, sometimes that kind of knowledge of what your competitors are doing is important as well.

I think he understands that there is upside by driving costs more aggressively and creating more headroom to either deliver better results or reinvest into the marketplace. So I know Dirk quite well and I'm -- I guess he will have a fast ramp-up and no complete or big differences in terms of the strategic perspective of the company.

And there's also a transition phase that has -- where he will -- of 5 months where he's -- 4 weeks, sorry, where he will work together with Niek Jan, but he knows the business extremely well. So moving on to Russell at New Street, please.

I'm going to try another one on the U.S., sorry if that's okay? Given obviously enormous synergies and sort of overwhelming logic to a deal, can you talk about what the kind of roadblocks are or issues or impediments or kind of why things are taking so long?

And then the second one is just on FTTH in Germany. Has there been any change in your thinking, please, about future investments or maybe anything to do with the timing of government subsidies or when we should think about some CapEx on FTTH coming through?

Yes, I'm going to start on FTTH in Germany. Basically for us, it's more or less 2 phases we're looking at, time-wise, one is we're going to finalize our vectoring rollout we're doing right now, also now addressing the Nahbereich, those areas where we unfortunately had some delays because of some long regulatory discussions driven by the EU.

But basically we're going to finalize that. As we all said, but end of '18, mid of '19, I guess we will deliver an around 80% coverage in the country with vectoring and then add a super vectoring feature to it by mid of next year.

So end of '18, mid- of '19, we should have an 80-plus percent coverage with 250 megabits which is I believe is a very good network infrastructure basis we're having here. So after finalizing that rollout, obviously, if you look at our current CapEx envelope, there is €1.5 billion incorporated just for that rollout of the vectoring infrastructure right now.

And that becomes available and for sure, we're going to reinvest that for -- into the next phase of rolling out infrastructure. Is it purely FTTH?

For sure not. You know our position.

We believe it is not about the right technology. There's no one right technology.

It is about finding the right mix of technologies depending on what competitive situation is, what willingness of customers to pay is and that's how we've done the rollout last years. And I think that's what we're going to do in the future as well.

But yes, for sure, from -- without having given any guidance from '19 onwards on that one, for sure, you would see a much larger extend of CapEx going into FTTH as well from '19 onwards. That's for sure.

On the U.S. roadblocks.

Again, it's -- the only perspective I can give is a conceptual one. The -- what you want to make sure is if you move forward and if you leave your stand-alone position which is an extremely strong one, worked hard 5 years to get to that very strong position.

But if you leave that and move into a deal, you want to make sure that you get more out of it. And it doesn't mean there are a lot of elements in the whole game.

So for instance, how sure you can be to get the synergies you have on a piece of paper really out? So looking at a practical example, MetroPCS, T-Mobile U.S., preconditions were very good to get to those kind of synergies.

And I think, John and the whole team has done an amazing job to do the retirement of the network infrastructure to move into one infrastructure, not losing customers on that way and so on and so on. So it depends not only on potential synergy numbers you might have on the paper, but also the A, capabilities and B, the difficulties you have to integrate and decommission and what you need to do in those type of deals.

And that's part of and should be part conceptually of any consideration. And as always, without giving any preference for that, a mobile-to-mobile merger is not only conceptually, but also by doing it the easiest way of creating synergies in the combination.

It's more difficult to create a business case by a cable or fixed-line mobile combination. We know that from Europe, but the case is still there.

So it is -- as we're always saying, we don't prefer the one or the other. It depends very much on the concrete situation and I think that's all I can say to it.

Yes. Thank you.

And with that, we move on to Andrew at Goldman. Andrew?

I have two questions. One on broadband growth and one on -- another question on network translation.

But on the broadband growth, you signaled kind of lack of -- an understandable lack of confidence in achieving your broadband CAGR. And so, just -- this really sounds a little dramatic, but I think, but what went wrong with broadband?

Is it an implementation issue? It seems that hard IP migration and other things have just got in the way or you've learned something more structurally about the growth outlook in fixed?

And then, secondly, on the network translation side. I mean, you mentioned quite a few times the €1.2 billion of savings that you set a couple of -- a few years back.

But I wonder if you could talk about whether that expectation has gone up? Or is it at least in the ascendancy with the emergence of new technologies and customer behavior.

So any update on the key new technologies or consumer behaviors that you would pick out as raising the scope of network transformation benefits would be great.

I'm going to start with the broadband growth. I think, first of all, let me state clearly that the overall CAGR for the total service revenues in Germany has not changed.

So it's like always; you have a mix of elements. One is your retail broadband and you have the wholesale part.

You have the mobile part and if you mix all that and come to the total sum of it, that hasn't changed. The inner mix has changed a little bit.

As you rightfully said, maybe -- and as I indicated, it's very challenging to achieve the 2%, maybe some upside on mobile; that's why the mix is still -- is a different one but total sum is then still intact. On the broadband growth side, basically it's 2 things.

It's twofold. One is the IP migration element.

And the way we look at it is there's still 6 quarters to go until end of '18 to get it done. And as I said previously, we need to work on mitigation measures to find better solutions set.

That's one point. The second one is, obviously, also a little bit of -- a lot of promotional activities the last 12 or 18 months.

And our offer to have the first 12 months for 995 and just try the new infrastructure, obviously, is part of that promotional elements, so that has been a headwind as well. Is that a structural problem?

I don't consider that as a structural problem as long as our turnaround into a positive territory of total service revenue is intact and that's true. So that's the first one.

The network transformation, when we did the calculation of the €1.2 billion, basically what has changed is, again, here also the mix of the contribution a little bit. When we announced it, it was 700, Europe 500 -- sorry 700, Germany, 500 Europe.

It was -- the larger chunk was by the customer service and field service. And a little bit like also 70-30 type of split was network infrastructure.

So the inner mix has changed a little bit of the contributed elements, but overall, no big change in those countries where we've -- where we're more advanced for the big ones like Germany. We've seen also that the effects are kicking in and that we're seeing those effects.

And it's basically the following elements, one is IP migration, that is network speaks one language. One is -- very important one is the virtualization of the network infrastructure.

So concepts like BNG or TeraStream, it's all concepts we're using to retire a lot of hardware you have in the network infrastructure. And to make sure that you have less components, less service in the infrastructure and that the software moves into the cloud which not only reduces your cost but increases significantly the agility you have, but also the way you can serve the customers.

I'll give you a practical example, for those areas, for instance in Germany, where we're using BNG infrastructure which is that technology way of virtualizing. What you can do is, you move your home in a fixed line.

We don't have to call customer service anymore. You log off, you pull the plug off your router, you put in your new home, you log in, that's it.

That is a very different experience not only cost wise, but also from a customer's point of view than the ones we had before. So it is -- IP migration is one part which is a necessity to have.

It is the virtualization of the infrastructure which is also part of the digitization we're doing. It is the rollout of the access infrastructure for sure and then there's another part which is those Pan-Net infrastructure where we put product platforms.

We have country by country into -- in one shape and consolidate them and just provide a service like TV out of 2 platforms instead of 10 platforms across the group. So those are the components and what has changed basically is the inner mix of the contributors.

But the vicinity of the savings are completely intact. And as I said, we see the results step-by-step also kicking in.

Very good. So next is Mathieu at Barclays.

Mathieu?

I have two questions, please. First, with regards to 5G and the evolution of mobile network.

Some of your competitors don't believe that small cells will be needed in Europe for a while and not for a number of years. Curious to know what is your view on that, if you actually believe something differently?

And second, with regards to IP migration, I'll focus on the costs. So you mentioned during the call that the peak costs were '16, but actually they're not declining that much in '17.

I guess maybe 6 months ago, you would have expected '17 to come down from '16. Yet obviously, your guidance is there and looking forward.

So what have been the offsetting factors, so far, on what you expect for H2 against the fact that IP migration costs are not coming down?

Yes. First of all, thank you, Matthew.

For the IP migration, basically what has changed during our initial view on the IP migration is basically two things, one is the speed we can transform and move the B -- the big B2B complex, B2B customers on a -- into a purely IP fashion and shape. That's one part.

And related to that, obviously, the IT and the product delivery supporting that. Knowing that this is the case and as I said, on the other hand, the positive is on the mass migration.

I think that's -- it's quite clear how it will be done and we're very confident by the end of 2018 we will have it done. I think knowing that, we've been clear about '16 and '17 but also been clear about there will a decline in '18 and we'll see, as I said, further cost reduction on the German side also related to that in '18 which will support the EBITDA.

So I think that message hopefully was quite clear. On 5G, I think there is -- what we're trying to do is first of all, A, leading the -- some of the initiatives on 5G in terms of specification and standardization to understand in a perfect manner where value sits and what needs to be done in 5G.

We're cooperating with those guys who are advanced like for instance, the SKT, the South Korean Telecom guys who are -- who will deliver a pre-version of 5G, I think, next year. So to understand better what's really going on in 5G and mobile network because I believe there's a lot of past there as well -- out there as well around that 5G thing.

I think what's obviously and what's clear looking at our European incumbent situation, that we have a lot of assets which will support a densification in terms of small cell, potential densification in terms of small cells. I'll give you an example -- and as you know, in Germany, extremely dense fiber infrastructure we're rolling out right now which is very important for small cells, that they have good fiber backhauling infrastructure.

Every -- not every, but nearly every street cabinet will be connected with a fiber connectivity that is in total 350,000 in the country. So you have that dense infrastructure.

You have also dense infrastructure of not only where fiber is available but also where power supply is available. So there is a lot of assets being available.

From a fixed-line perspective, you can make use of that densification of the infrastructure. And for us, small cells is one element we're going to use to improve network further on and then keep the differentiation on the network side.

By the way, that holds true for Europe with the now incumbent situation. Fortunately, also for the U.S.

where the fiber connectivity to small cells is easier to achieve also without having a fixed-line infrastructure. So yes, we're going to go for it and show the customers that there is upside for them in there.

And I think, in volume terms, I think, you've heard Thomas earlier in the call that we have commissioned 25,000 small cells in the U.S. coming on top of our 13,000 existing cells.

And that's sort of hedged probably in 2-year perspective although we often find our U.S. guys a bit faster doing stuff.

And surely, we support it as much as we can. And in terms of -- and that's I guess we'll continue.

And in Europe, in Germany, I think you should expect small cells deployments to slowly ramp up and move into these kind of volumes. Probably, more early in the next decade but since we haven't talked about this -- our CapEx beyond '18, we should not say too much right now.

But it's a good subject, I guess, for our Capital Markets Day next year, that you have heard about it today. So let me move on to Sam at Exane.

Sam, can we please have your questions?

Just two quick questions. Firstly, on the all-IP migration.

I'm just wondering, you seem frustrated by this law that forces you to go and cancel the contract. Any thought of that changing at all you think?

And I guess the benefits you've talked about from all IP simply from switching off legacy platforms. Is that something we can expect to hear about next May as well?

Secondly, just on kind of bond redemptions. I think you said there was about €200 million in cash payments, sort of premiums this year.

Is there scope for more to do later this year and next year? Or should we expect a nice boost of free cash flow next year as those costs kind of rollover?

Yes, first of all on the frustration part on the all-IP, unfortunately, that will not change. The court was a bit high.

I think it was Federal high court. So there is very little reason to think about a change here.

And even if it's in the interest of a customer, sometimes, the legal environment doesn't support that. So we will need to find better ways.

That's our task now within that environment, but I guess challenging that environment and changing it -- that is not -- will not really help. On the €200 million effect, it depends on -- obviously, there is still opportunity to optimize the financing situation in the U.S.

Whether we do that or not is subject to a discussion we'll have with our U.S. colleagues and team and see then whether we do it in '18 again and then we will have a rollover or not.

Let's see; that is -- that's still open.

But either way, it will, of course, help us with the free cash flow next year in more than one way unless we have another one-off year. But I think it's probably good to know that at the half-year stage we had €600 million more free cash flow than last year.

If you take the €400 million that we reported plus the €200 million that we invested in lower interest cost going forward. So I think that's helpful.

And the next question --

Wait a second Hannes. There is one element I want to touched on -- upon and it is, you guys should feel comfortable that we deliver what we tell you in terms of the 10% CAGR on the free cash flow and the guidance we've given for this year, the €5.5 billion.

And there's no doubt we will deliver that. And the same holds true for 2018.

But you should not get carried away. Our perspective in terms of -- our intention is to create headroom above the 10%.

And whenever we create that headroom above the 10%, we're going to reinvest it to make sure that we can keep going with the free cash flow growth. So let me state that very clearly, you should expect that we deliver what we promised, but whenever we create headroom, we're going to invest it into the future of that business.

And so, with that, let's move to Justin. Justin at Credit Suisse.

Can we have...

Yes, just follow-up questions, please. So on the all-IP, does it get worse before it gets better?

It feels like you got sort of tougher customers left on the B2C side. So we're going to see a pickup in line loss continuing through this year and next year because of that.

Secondly, your new offer with Sky on TV in Germany, is that some -- you've put it up selling at cost or in fact selling it below cost. And do you think now that -- actually, you can start winning back customers from cable, particularly, some of those STU customers on cable, pairing it -- paying a fair amount for TV.

And then thirdly, on mobile. The -- obviously, the way that StreamOn works is you're selling it through M and L tariffs.

Can you show us how the mix of your T-Mobile customers, contract customers, the mix that are on M and L, has that changed between Q2, Q1? Is it starting to go up?

We do not disclose the concrete numbers on M and L, but as I said, there is an increase on the L tariff by that streamOn -- and M and L, by the way, both from S to M from M to L, so you see that increasing -- you see the numbers increasing there as we have expected. On the IP, I think the worst part about the IP we're making right now is obviously other than the cost which is expected element and there is that churn element.

And I think that vicinity of high-20s, mid-30s, that's where I believe we will see for potential -- if we don't find better mitigation we will see for next quarter's impact. And that's why I said, we're focusing on mitigating those numbers.

We will not get them completely off the table, but every customer counts. So we're focused on bringing them down.

But it's not like -- they will not double. So that's not the case.

On the Sky offer, your question on the Sky offer. Let me be clear what you get and what you don't get with the new offer.

The new offer is basically a basket by Bundesliga. It's Eishockey-Bundesliga.

It is third foot soccer Bundesliga, then it is selected games of the women's soccer. And that's all products we have available anyway on our site where we have the content available.

What we add from the Sky, from Sky is Handball-Bundesliga. We add the conference -- the live conference on Bundesliga and the live conference on Champions League.

So it's not all the games or every game live. It is the conference for Champions League and for the Bundesliga Live.

So it's kind of, as we believe, an extremely attractive entry-level product, so to say. If you want have the Full Monty, you can get it, you can get it via us and Sky and then the corporation.

And you can get the Full Monty where you can see all the games live. But we believe there is a big chunk of customers at a relevant segment which would love for -- to spend for like €10 a month to see that kind of content.

And I guess we will find out next 12 months in the first phase and then the free-to-air on Champions League will end, Hannes, I think May next year, is that right? So from May, June next year onwards, you will not see any Champions League anymore free-to-air.

So that will give us, I think, a second step, another boost. So that's how we think about it.

And obviously, yes, it is about attracting customers, also new customers with that kind of content.

Yes. And we think it's a good deal from a commercial logic and it's better, I think, under this German circumstances than models that we have seen and certain other markets in Europe.

And a great way to different -- achieve differentiation without spending an unreasonable amount of money. So the final question today is from Wolfgang Specht at Bankhaus Lampe, please.

Two questions from my side. One, I read in a new channel that you are about to receive a compensation payment from BT.

Can you confirm that? I read an amount of around €200 million to be paid out in Q3.

And then, the continuum question would be, is this included in your full year free cash flow guidance? That's it from my side.

Yes. Thank you, Wolfgang, for the question.

I think what has happened is BT has reached settlements with us and Orange and in repair and respect of any warranty claims under the 2015 EEA acquisition agreement arising from issues that were previously announced by BT regarding the global services operations in Italy. And I think we're pleased that we have found a consensual solution to that retroactive matter.

It's already -- it feels like 100 years ago, but I think it's only 2 years ago. And under that agreement, BT will make that payment of GBP 180 million which is roughly €200 million.

And this settlement concerning that retroactive matter, of course, from all point of view, has no impact on the future relationship between the two companies. It's just related to what we had there in the SBA.

How does that impact our free cash flow? It doesn't impact our free cash flow because it's related to the M&A transaction and those elements are not part of our free cash flow or reported free cash flow.

And so the only thing which is impacted positively is the net debt number and the ROCE number, basically.

Okay. Thank you, Thomas.

So with that, the conference is now coming to an end. And I would like to thank you all for your kind attention and questions.

And if there are more questions that we haven't addressed today, then I would like to ask you to contact our Investor Relations Department. And with that, I hand back to the operator.