Good morning, ladies and gentlemen, and welcome to Orange's Third Quarter 2020 Results Conference Call. The call will be hosted by Mr. Ramon Fernandez, Deputy CEO, Finance, Performance and Development, with members of Orange's Executive Committee for the Q&A session that will start after the presentation. Thank you, and let me now hand it over to Mr. Ramon Fernandez. Start the Q&A.
Thank you. Good morning. Good morning to all, and welcome to the presentation of our Q3 2020 results. I am going to present the main highlights of Q3 before going to the business review. At the end of the presentation, I'll do a brief focus on two topics which have been of growing interest to the markets: wholesale fiber co-financing and our infrastructure strategy. Slide four presents the key achievements of the quarter and the resilience of our activities. As a result of our excellent commercial performance this quarter in our main countries, we now serve, at group level, nearly 11 million convergent customers and around 9 million very high broadband customers out of 46 million connectable homes. The increase of 22% in our VHBB customer base is notably thanks to France and Poland, which both posted the highest ever quarters for fiber net adds.
In mobile, Orange won in France the highest number of 5G frequency blocks with a total of 90 MHz, securing our network premium leadership at a reasonable price. After Romania and Poland, our 5G-ready mobile offers have been launched in France and Spain. Orange Bank now serves 1.1 million customers and is successfully pursuing its value strategy with a record in volumes of sales of premium cards this quarter and nearly 60% of new customers in Q3 in France subscribing to paying offers. Last but not least, Orange has successfully issued a nine-year EUR 500 million inaugural sustainability bond with a 0.125% coupon, supporting its social and environmental commitments. Turning to slide five, despite the crisis, our top line was back to growth this quarter at +0.8%, fueled by a solid performance in France and Africa and Middle East.
Europe, which now includes Spain and enterprise, are still under pressure but on a better trend. The group EBITDA trend improved significantly this quarter at -0.4% compared to -1.7% in Q2, despite remaining negative effects from the crisis of around EUR 150 million. Successfully pursuing the monetization of our fiber network in France, we received, once again this quarter, co-financing from our competitors. Over the first nine months of 2020, EBITDA declined slightly by 0.6%, and eCapEx decreased by 6.3% while we accelerated the rollout of both our very high fixed and mobile broadband networks. This decrease is mainly due to eCapEx deduction coming from co-financing. To conclude on our key financials, at the end of September 2020, our year-to-date EBITDA minus eCapEx ratio has improved by 6.3%, and as you can imagine, our organic cash flow is also well on track.
Now, let's turn to our business review, starting with France on slide seven. In Q3, total revenues increased by 3.1%, driven by co-financing revenues for construction in PIN areas and the resilience of our retail services, despite the negative effects of the crisis on roaming. In order to assess the underlying performance, let's analyze our retail activities, excluding the digital content offers. Retail services turnover, excluding PSTN, grew by 1.7% in Q3, driven by convergence, and retail services, including PSTN, were stable in Q3. Convergent services grew by 3.3%, driven by price increases and the increase in the number of lines per convergent offers. Convergent ARPO increased by EUR 1.2, reaching EUR 68.8. Mobile-only revenues decreased by 2.3%, mainly impacted by roaming and the decrease in prepaid offers targeted to international tourists.
Broadband-only services were up 2.7%, thanks to the growth of our customer base and the positive effect of premium pricing on newly migrated fiber customers. Equipment sales improved from -26.7% in Q2 to -9.8% in Q3, thanks to the reopening of our stores. iPhone 12 release this month should contribute to accelerate this trend. Finally, wholesale turnover grew strongly, with +10.5% after +11.7% in Q2, driven by the effect of co-financing received in mid-dense areas and by the growth in PIN construction revenues, more than compensating the decrease in national and visitor roaming. This was supported by an excellent performance on mobile and broadband, with a record quarter in fiber. On mobile, we delivered a very solid performance with 125,000 net adds, the best quarter since Q2 2018, driven by both SOSH and Open, while churn is normalizing.
Following the reopening of our shops, the Orange brand has strongly recovered during Q3, fueled by convergence and high-end offers. On fixed, we recorded 95,000 broadband net adds, driven by 360,000 fiber net adds. A great performance obtained throughout the quarter, of which 53% are new customers, consolidating our leadership position with 4.1 million fiber customers and 20.9 million connectable homes. Let's now turn to our seven-country Europe segment, which from now on includes Spain. Even if the pandemia has continued to limit the free circulation of people, the commercial performance of the seven-country Europe segment significantly turned around in Q3 for all product lines. We have 221,000 mobile contract net adds, 96,000 fixed broadband net adds, of which 140,000 fiber lines. This remarkable performance is mainly the result of the revamping of our product lines implemented in most of our countries during the back-to-school period, traditionally supportive for business.
Total revenue remained under pressure at -3.7%, mainly because of a greater reduction in roaming this quarter, which accounted for more than half the total erosion year on year. Yet, the total revenue trend is improving by 140 basis points versus Q2 because of better equipment sales themselves linked to better service net adds. In pro forma previous Europe perimeter, excluding Spain, the trend in total revenues improved. It would be -2%, nearly stabilizing, excluding roaming. It is also worth noting that Orange Poland set a new record in quarterly fiber net adds, up 54,000, and that Orange Belgium passed the 300,000 milestone for cable customers only four years after it launched the offer.
If you turn to Spain, slide 10, you have a total revenue trend that was better oriented at -5.6% in Q3 compared to -6.8% in Q2, despite the negative effect of the crisis on roaming. The decline of retail services revenues at -7.7% in Q3 is mainly due to volume loss linked to our weak low-end presence in the past. In order to improve these results, we revamped all our portfolio in H1. Our new commercial strategy is already demonstrating its effectiveness in the period of the back-to-school and the launch of the football season, with a return to positive net adds everywhere: in fixed broadband, in mobile, and in pay-TV. Q3 was the first quarter of growth in our convergent customer base since Q1 2018. Strengthening our presence in the low-end convergent segment resulted automatically in the dilution of our convergent ARPO in Q3, -2.4%.
It is important to highlight that the ARPO of the Orange brand grew in Q3 and that 80% of the customers of our new low-end convergent offers are new customers for the group. In Spain, we managed to significantly improve our commercial trajectory in this highly competitive market, and this will pave the way to better financial results, which will materialize in the medium term. Let's now turn to Africa and Middle East, with revenue in Q3 returning to our target level, increasing by 5.1%, driven by retail services at +7.1%, with data reaching almost 31 million 4G customers, with revenues up 19%. Orange Money, with an active customer base exceeding 20 million and an accelerating dynamic delivering +27% revenue growth, the highest level over the last 12 months, which shows an ongoing appetite for mobile financial services.
Capitalizing on this strong performance, we successfully launched Orange Bank Africa last July. Finally, fixed broadband, with almost 1.6 million customers and revenues up 26%. Looking at the commercial KPIs, the mobile client base is increasing by 2.3% to 126 million customers, and the customer base quality keeps improving, as reflected by an increase of two percentage points in the charge base rate and the reduction of one percentage point in mobile prepaid churn. From a geographical perspective, revenue grew in 12 countries, including six at double-digit pace, with top contributors to total revenues being the Ivory Coast cluster, up by more than 13%, Egypt more than 8%, and the Senegal cluster. Finally, the enterprise segment posted a Q3 top-line decline of 1.7%, improving the trend compared to Q2, where it was - 3.3%, despite the COVID crisis, which continues to have a significant impact on the B2B business in 2020.
The top-line performance is explained by a decline in mobile, which is fully linked to a -71% revenue decrease in roaming, a decline in fixed-only services, with voice decreasing by 3.4%, but still remaining above the pre-COVID trend thanks to our voice-over IP offerings and our collaboration solutions, and data decreasing by 1.3%, still impacted by the cancellation of events. Finally, IT and IS has almost returned to stability at -0.5%. If you exclude equipment, IT and IS was growing by 3.4%. In this context, cybersecurity revenues grew by 8% and cloud by 5% in the first nine months of the year. If you exclude a significant one-off contract from last year, cloud was even growing by 14%. All in all, the segment profitability has significantly improved compared to Q2. We know we do not communicate on EBITDA in Q3, but significant improvement.
We are continuing our partnership strategy with, for instance, an agreement with KDDI to provide a customized IoT platform for Japanese automotive manufacturers, Toyota and Mazda. We also launched a digital transformation pilot with Schneider Electric for the first factory in Industry 4.0 powered by 5G. Now, let's take a few minutes to focus on the execution of our infrastructure strategy. I'm going to address two key questions that we've been asked quite often in recent months. First, what about co-financing FTTH monetization in France? Second, how do we execute our infrastructure strategy as disclosed last December? These important questions are part of a broader infrastructure strategy, which basically defines how we will deploy controlled or co-controlled infrastructures while maximizing their value and return at group level. This strategy is based on three key pillars.
First, find the smartest way to finance the deployment, which depends on the asset class considered. Second, capture growth thanks to optimized network management. Third, manage some of our infrastructures as an independent class of assets when we consider it would create more value than keeping these assets under the umbrella of MNOs. The operations I will talk about briefly will allow to reveal this value. First, co-financing and fiber in France. I'd like to remind you that the fiber business case is built upon two pillars: monetization through our retail and wholesale activities. On wholesale, operators have two options: co-financing and line rental under conditions which are set by the regulator.
The arbitrage between rental or co-financing by French operators implies the following for Orange as a wholesaler: rental choice offers higher monthly rental fees, meaning higher recurring EBITDA contribution, while co-financing implies lower monthly recurring fees but higher upfront amount, externalizing CapEx and commercial risk, thus improving our ROCE position. Now, to give you an important indication, the accumulated co-financing proceeds received between 2009 and this Q3 2020 amounted to EUR 2.4 billion. These proceeds represent less than 50% of total long-term estimated potential based on our competitors' current broadband market share. This is my first message. What we've got is less than 50% of what we can get. Second message: the French fixed wholesale revenues trajectory for the period 2019 to 2023, given during our last capital market day in December 2019, is unchanged. This is the second message.
Third message: we do not expect any shift in the retail competitive dynamics, as the competitive situation is not new to Orange. Our competitors have been able to access our ADSL customers through full unbundling since 2005, and yet we managed to stabilize our market share after a few years thanks to Orange retail and service capillarity, brand advantage, and long-standing local presence. All these elements contribute to our FTTH profitability target of an incremental IRR of around two times the cost of capital in France. This was my first point. Second point: let's take a look at the execution of our, let's say, the equity part of our infrastructure strategy disclosed last December, on which we are accelerating, turning to slide 15. Fifteen is European Tower.
Carve-out work is ongoing under the leadership of a dedicated project organization at group level, including teams from France and Spain operations and external advisors. The design phase should end by February 2021, a few months ahead. This will allow us to share a comprehensive view of the design of our Tower Co for France and Spain at the 2020 annual results, including the scope of the assets transferred, governance, and detailed economics based on negotiated MSAs. This project is not just about a carve-out. Our goal is to move one step further towards converting our mobile passive infrastructures into a real business to capture future market growth organically or inorganically.
In this perspective, conscious of the value of these assets, Orange is willing to keep control of its European passive mobile assets while managing them as an independent business, enabling commercial neutrality and operational efficiency, including through arm's-length contractual relationship with MNOs, dedicated management accountable for performance, appropriate governance, and ad hoc reporting to disclose performance and strategy. As part of this strategy, we are also looking seriously at flexibility on the capital structure of the European Tower Co to accompany its growth while keeping control over it. Second, slide 16. With the fiber assets, Orange is creating ad hoc vehicles to strengthen its already strong position in the European landscape. Regarding France, we talked to you in December about our Orange Concession project, and we now have made an important step forward. We are currently selecting a partner to acquire 50% of Orange Concession.
We have already received high-quality non-binding offers mid-October, and we will choose a partner in H1 2021. In Poland, we are advancing quickly with the creation of a Fiber Co targeting 2.4 million households. We are looking for a partner to acquire 50% of this Fiber Co a competitive process has been launched this summer, and we expect non-binding offers by mid-November. We expect to choose a financial partner by the end of H1 2021. Let's now wrap up today's presentation with a summary of our guidance. With regard to 2020, I'm confirming that we anticipate a slight decline in EBITDA in 2020, around -1%, including all COVID-related impacts. Concerning eCapEx, and given the delays observed so far, eCapEx will decrease in 2020, offsetting the erosion in EBITDA. As a result, EBITDA less eCapEx will be stable in 2020.
Our 2020 organic cash flow guidance is unchanged, with a commitment to exceed EUR 2.3 billion. With regards to our dividend policy, confident about our organic cash flow guidance, the board of directors of Orange yesterday supported the return of the dividend to EUR 0.70 per share for the year 2020. The final decision will be taken in the light of the final year results. We will pay an interim dividend of EUR 0.40 on December 9, representing an increase of EUR 0.10 compared to the amount initially planned. This concludes my presentation, and with all my colleagues, we are ready to answer your questions. Thank you.
Thank you, ladies and gentlemen. We will now start the Q&A session. In the interest of time, please limit your questions from one to two at a time.
In order to ask the floor, please press star one on your telephone keypad. Please ensure that the mute function on your telephone is switched off to allow your signal to reach our equipment. If we do not have time to take your question, then please contact our investor relations team after the call. If you find that your question has already been answered, you may remove yourself from the queue by pressing star two. Please press star one to ask a question. We'll now take our first question from Nicolas Cote-Colisson from HSBC. Please go ahead. Your line is open.
Thank you. Good morning. Two questions, please. First is on the COVID impact and the new lockdown measures. Can you explain how the pandemic is impacting your CapEx because you are accelerating your fiber rollout at the same time?
Where do you get some CapEx flexibility, and how do you see the risk of a catch-up next year taking into account the transfer of assets that may change the equation too? On the co-investment, maybe can you indicate the proportion of the EUR 2.4 billion coming from medium-dense area? Because if it's the bulk of it, then it would indicate that about 50% of the lines you built have been co-invested. I was wondering if it was a fair assumption because you also mentioned that only two of your three competitors are very active in that area. I'm just trying to square the circle around this EUR 2.4 billion. Thank you.
Okay. Maybe I'll start, and Jérôme Barré will be completing the answer on the second question. Thank you, Nicolas, for this question.
On the COVID impact and the CapEx perspective, if you look at it on a basis of, let's say, 2020 and 2021, it's obvious that 2020 is going to be significantly lower than initially expected. We should end the year of 2020 around EUR 7.1 billion. This is significantly less than what we had initially guided for because we had said initially that we would do roughly EUR 200 million more than the EUR 7.2 billion of 2019. You see, EUR 7.1 billion should be the outcome. It was even EUR 7.3 billion, in fact. This is 2020, ending the year around EUR 7.1 billion. If you take 2021, there will be a catch-up in all the countries where there were some delays in the rollout of network. As a whole, if you take the two years, you should end to the same level than initially expected, which is roughly somewhere between EUR 14.7 billion-EUR 14.8 billion.
If you add 2020, 2021, you would have roughly the same figure because there would be this catch-up. I hope it's clear. On the second question, I'm handing the floor to Jérôme.
Thanks a lot. Good morning, everybody. Your first question was about the repartition between the middle-dense and the very-dense area. Roughly, it's 70% for the middle-dense and 30% for the very-dense area. If I may add some details about that, you see, we say that in the private area, so far, the global amount of the global number of lines which have been co-financed represents 80% of the total fiber accesses of our competitors, of course, of the networks which have been deployed by Orange. That is to say that 20% are still rental lines. As you mentioned, there is one competitor who doesn't co-finance.
Why do we say that the potential is still a little below that 50%, still to be co-financed? It can be explained by a double effect. First, a significant part of the private area is still to be deployed, so the surface will increase. The second point is that a large number of customers are still to be migrated from copper to fiber. I mentioned this 20% of customers who are under rental scheme, but there are also still a lot of customers who are still on copper lines. When you add together those two elements, the increase of surface of deployment, the footprint of deployment, plus the final migration from copper to fiber, it explains this move to 50%.
That is very clear. Thank you very much.
Thank you. We'll now take our next question from Roshan Ranjit from Deutsche Bank. Please go ahead.
Great. Good morning. Thank you for the questions. Two for me, please. Just focusing on the domestic mobile side. We saw some very good KPIs supporting retail trends in France. I think you mentioned a mix of both SOSH and the higher-end brands. Now, when I look at the KPI file, we have seen a nice tick-up in the SOSH ads. If I think about some of the pricing adjustments we've seen in the market more recently, I think you have been a bit more active on SOSH. Can I just check that you are still seeing the majority of your ads coming in at the more premium end of the market in France, and there's no kind of concern given the pandemic in recent months that people are taking a lower-end plan?
Secondly, I guess fitting in with your infrastructure timeframe, I think recently Arcep increased the number of sites as part of the new deal. Thinking back to last year, you did sign a build-to-sue program with a third party. Can I think how is it possible to get an idea of how you will be dealing with these increased sites in the rural areas? Will that be part of the Tower Co, or will that be scoped for another build-to-suit? Thank you.
Thanks. I'm going to hand over to Fabienne. I'm just going to say one thing on the first question, which is that, and I'm paying tribute to Fabienne and her teams, right? She will not be daring to be so proud. I'm saying that in France, we took the risk to trigger the move upwards with prices with SOSH starting this spring and summer.
There is a much better pricing environment now in France. I think that the moves we have made, including with SOSH, have been absolutely instrumental in going into this better environment. Fabienne, the floor to you.
Yes. Good morning. On the mobile, you're right. We deliver a real solid performance on Net adds. It's the best quarter recorded since Q2 2018. This performance is driven by both SOSH and Open. This is very important. We are very satisfied by the mix because we observed in Q3 a good performance on Open thanks to the reopening of stores. The digitalization effort we made, we observed a strong dynamic on Orange brand, and especially on Open, and that fueled the mobile performance.
It's a very good performance, and we are really confident that month after month, the brand Orange fueled by Open and the high-end offers will be more and more efficient. It's a good performance. The high-end offers are around 44% of the Net adds. It's a really good performance. As explained by Ramon, we observed a second point very significant in this quarter, a market repair. I think we can use this word, especially on the mobile. We observed that at the end of Q2 and confirmed in Q3. This is a really good point because we observed an encouraging price recovery from all competitors, both on low-end and high-end markets.
This is very significant for Orange because we can maintain and pursue the strategy we have to push the premium as we made the 8th of October 2020 when we launched a new plan revamped on the mobile with the same strategy focused on more for more. It is exactly what we did, and it is exactly what we pursue in the future. It is really well-oriented. For your second question regarding the new deal and the build-to-suit, I do not understand exactly the connection because regarding the build-to-suit, we intend to keep current momentum. We are stable, and there is no acceleration in the build-to-suit program we designed. There is no connection to the existing new deal agreement. The new deal agreement has been fixed by the government. There is no discussion about it currently. The new deal is ongoing, and we deliver the commitment we took, especially in June 2020.
We were the only one player to deliver the commitment with the 104 new mobile sites delivered as expected by the government. We don't need to have an acceleration on the rural area on the mobile because if you accumulate what we do on momentum plus the new deal agreement site mobile expected, we are on line to be exactly what we have to be and what the government wants we are. I don't think we have a question or an issue about this point.
Yes. Roshan . There is no, I accept that there is no request of increasing anything on the new deal agreement. It's totally on track.
Okay. Sorry. Sorry. I just checked. The new deal is standard at 20,000 sites, and I thought it had gone up to 30,000.
No, no. Nothing changed.
No change. No.
Great. Okay. Thank you.
Thank you. Our next question comes from Jakob Bluestone from Credit Suisse. Please go ahead.
Hi. Thank you very much for taking the question. I was just wondering if you could comment a little bit around the dividend decision to increase the interim. What's behind that? Thank you.
Jakob, what's behind that is the very strong conviction that we have a capacity to pay this dividend of EUR 0.70, and the board has supported this yesterday based on the nine first months of the year. We have seen on the basis of Q2 and Q3 that the resilience of our business puts us in a very solid position to pay the EUR 0.70 for 2020.
Probably adding EUR 0.10 to what will be paid in December, in a way, compensates the EUR 0.20 reduction which was decided on the 2019 dividend and can give you extra confidence that the EUR 0.70 will be paid. Very good commercial performance, very good operation performance, and very good execution on what we have decided to do with our infrastructure's plan. Cash flow is there. We are able to pay, and this is confirmed by the EUR 0.40 which will be paid in December.
Thank you very much.
Thank you. We'll now take our next question from Nayab Amjad from Citi. Please go ahead.
Thank you for taking my question. One question on France retail. Would you consider backbook pricing increases given that the SOSH entry price level now is EUR 16? I mean, your competitors have been already doing that.
Would Orange consider something like that? On Spain infrastructure, the ex-Poland CEO, who is now the CEO in Spain, started the monetization of fiber assets in Poland. Do you think it will be possible to consider monetizing the fiber assets in Spain, or would you endeavor to maintain control? Thank you.
Fabienne will take the question, of course, on France. Jean-François Fallacher, our new CEO in Spain, is also with us and will take the second question.
Yes. On the mobile, it's exactly what I explained just a few minutes ago. We observe an encouraging price recovery movement. On the mobile, the point of the low-end offers pricing is currently around EUR 15. If you remember, one year ago, we were around EUR 10, EUR 12 , EUR 10, EUR 12. So, it's a significant move.
In this case, that sustains all the pricing premium strategy we can push. The second point interesting to highlight is that the first 5G tariff and plan launched by competitors as Orange, but not only, indicate the wish of the old competitors to sustain a premium in the 5G too. In this case and in this context, we push to momentum on SOSH, and we increase our price on SOSH. We are around EUR 16, and we launched last in October, three weeks ago, a new plan, full new plan. We revamped our offers with the strategy of more for more. We decided one year ago, and we are very satisfied by the first week. We have a very good momentum in this new plan we launched.
It's around EUR 2 or EUR 3 or EUR 4 more, but still with the same strategy, more for more, more data, and SIM card additional. We are confident more than in the past on the very positive movement we can have on the mobile market, French mobile market. We are really confident. I think we are waiting for this confident position that we are now.
Thank you, Fabienne. Jean-François?
Yes. Hello, everyone. Jean-François Fallacher from Madrid. As concerns our fiber infrastructure, the question was, are we having some plans to actually dispose it or monetize it through fiber contact of deals, if I understood well? The answer is that, obviously, our fiber infrastructure in the Spanish market is definitely for us a critical asset with more than 14.2 million lines, actually, home paths , in Spain.
We are having, as you know, a project to increase this footprint by three additional million. Here in Spain, the decision has been made to do it under a leasing scheme with Santander, which is actually having the benefit of being a cheap way to get financing on further deployments, with us keeping the property and the ownership of this fiber at the end of the leasing contract. There is no such plan, I mean, to obviously seize this critical asset. On the contrary, we definitely believe more than ever, and you have all seen that through this COVID crisis, it has been shown that providing, obviously, the right internet access to consumers and businesses is critical in each of our countries. Basically, no such plan to build a FiberCo in Spain.
I remind that this is an asset that we are also heavily monetizing, not only to our retail customer base in Spain, with more than 4 million customers on this network, but also through a number of wholesale contracts, which can be ranging from bitstream access to some kind of co-financing schemes with some of our competitors. My answer is that, first, we want to keep the ownership and the control on this critical asset. Second, I mean, the team here and my predecessor have been very busy in monetizing, again, not only in retail, but also in wholesale, this critical asset.
Thank you, Jean-François. Thank you.
Thank you. We'll now take our next question from Frederic Boulan from Bank of America. Please go ahead. Your line is open.
Hi. Good morning. Two questions from my side. Firstly, on B2B, you seemed a bit more confident on the evolution of the business. If you could shed a bit of light on the EBITDA and the cost initiative you have here and what you're seeing from a competitive standpoint. Some of your competitors have been pushing harder on B2B. Do you see that in ongoing commercial activity? Secondly, on your asset portfolio, you have discussed a couple of angles today. If you can extend the commentary around potential sharing agreements in France or elsewhere, any potential IPOs you see could make sense in the portfolio. I think you mentioned cybersecurity recently. Maybe on towers at this stage, if you can clarify your preferred ownership structure, if I understand well, the consolidation is not on the table, even if it could create some value. Thank you.
Okay. Thank you, Frederic.
The B2B question is obviously a question which is of interest for both Orange Business Services and all the MNOs who have some enterprise business within their activities. Maybe we can concentrate the answer on Orange Business Services. If we want, we will be able to follow up with, for instance, Fabienne and the other colleagues around the table. Helmut Reisinger is going to take the first question.
Yeah. Thank you, Frederic, for this question. First of all, we are more confident, indeed, because the decrease in revenues is much less compared to the second quarter, with down now - 1.7%. Notably, this represents in absolute numbers about a EUR 33 million decrease. If you look into there, actually, EUR 20 million of this decrease is simply equipment resales. That is non-services-related revenues. About one-third is services-related.
On this part, if you just think about the fact what Ramon has already highlighted, that we have a major decrease still in B2B roaming income. This amounted to around EUR 18 million. The events that Ramon was referring to was, for example, the Tokyo Olympics. We are very proud to have signed this contract already a while ago, but the Tokyo Olympics did not take place. That is another big ticket in there. For the rest, I must say that our motors of engines of growth in terms of cloud services still solid double -digit. We believe that when we look also into our order intake, and we usually do not report on total order intake because we focus so much on new and get orders, which are orders for new services, either on new customers or existing customers.
Here, actually, the decrease, if we look to Q3, was only 3% compared to a strong order intake last year. We are quite confident that this trend continues. If it is true that the post-COVID world will be more connectivity, if it is true that it is more cloudified, it is going to be also more digital, and it is going to need more cybersecurity. For all of these growth trends, we have the right setup, which is one core business supported by four growth vectors, which is cloud, cyber defense, digital analytics, and—you have seen this in announcement of the new partnerships—smart mobility services to anticipate the 5G services that will come. I hope this gives you a bit of flavor. In profitability and cost measures, we continue to be extremely tight and disciplined on that.
Just to give you also a flavor, in our outside France activities already, we have an offshoring rate of labor which is close to 50%. Just to show you that we have not started this yesterday, but it is a continuous evolution that we are very much focused on to create value for this business.
Thank you very much, Helmut. Patrice is sitting next to me and tells me I am allowed to say that the EBITDA trend in Q3 is close to three times better than in H1, right? You can guess that H2 will be much better than H1. Helmut, we are authorized to say so to our friends. This was question one. Thank you, Helmut. On the second one, Frederic, around the asset portfolio management, which has really a number of elements to it.
I'm going to try and give a few elements, and of course, we'll see if we need to be more detailed. I would say that all the points you've listed about network sharing agreements, potential IPOs, towers, etc., our real priority is value creation and how we can support growth in the company looking ahead. This is the common theme. If you look, for instance, I'm talking under the control of Fabienne at network sharing. We have network sharing agreements nearly in all our European countries. We have extended them in the past when it was good for us. Of course, good also for others, but we are interested in what's good for us. It would be exactly the same for France.
If there is a case to go into a run-sharing arrangement, as it has been alluded to externally, we will do so, but uniquely only if there is a case. For the time being, no discussion has been started on this. First point. On IPOs, I would say, because here also, from time to time, we see some comments, an IPO is a means to an end. It is not an end in itself. If an IPO can be instrumental to support growth, to accelerate, we will consider IPOs. If not, there is no case. You have heard us talking about, for instance, Orange Africa and Middle East. This will be the framework to assess the situation. For cyber defense, and Hugues Foulon is also on the call, it is the same on cyber.
The top priority for us is to grow, and the teams have been extremely successful, including with some M&A, which has been very successful with SecureLink and SecureData. We are now close to EUR 700 million revenues in cyber. We want to continue to grow. Then we will see what are the options which help us to continue to go in this direction. Same then for towers. Towers, infrastructures, I hope I was clear enough when I tried to explain what is the framework, but you can see that our approach on towers is really to go in managing a business independently in order to capture growth. When I said flexibility, and this is on the slide, flexibility on capital structure to fund growth, when you talk about flexibility on capital structure, of course, you have an open view on different schemes.
This includes welcoming other shareholders, including potentially at some point through listing. You keep control. You keep control of the asset, which does not mean you are 100% in control. Okay? This is the general approach. I hope it is clear enough. If any of my colleagues want to complement this, of course, most welcome.
Yes. Thank you, Ramon Fernandez.
Thank you. We will take our next question from Abhilash Mohapatra from Berenberg. Please go ahead. Your line is open.
Yes. Good morning, and thanks for taking my question. I have got a slightly longer-term question around the wholesale business, please. I appreciate you are already giving us a lot of colors sort of by committing to this 2019 -2023 kind of revenue profile for the wholesale business. Just wondering, how should we be thinking about the revenue profile once the co-financing proceeds are out of the way?
Just related to that, I'm just wondering, other than corporate unbundling revenue, are there any other kinds of revenue streams that might be at risk over the long term that we should be keeping in mind? Thank you.
Thank you very much. Jérôme Barré will take this.
Okay. Ramon says that the French fixed wholesale revenue strategy for the next period, 2019-2023, given our CMD, last CMD, is unchanged. If I can give some details, I'll try to answer your question about that. First, concerning fiber, you have to take into account that we have several sources of revenues and wholesale in fiber. First, if I take the first case, which is a local loop, revenues, we get revenues, recurring revenues, co-financing revenues in the areas where we deploy fiber.
You must be conscious that in the areas where Orange doesn't deploy fiber, we get revenues as well because we monetize our infrastructure through ducts and hosting, and it represents all over this territory recurring revenues. In addition to the local loops, which is the last mile, and on the whole territory, both in the areas where we deploy fiber and areas where other operators deploy fiber, on the whole territory, we aim to increase our revenues on backhaul aggregation and activated services. You have to add altogether this local loop revenues of fiber, ducts and hosting revenues, and backhaul aggregated and activated services. On the other hand, that was the second question about fiber. Our objective is to increase the unbundling tariff. You have probably noticed that the first step has been carried out with a recent announcement coming from Arcep.
When you add the unbundling tariff itself and the commissioning tariffs, this represents an increase of EUR 0.30 next year. Our objective is to make these figures increase in the following years. As you can notice, we work both on the global revenues on the fiber and the revenues of copper, and that was the reason why we maintain our guidance. Thank you.
Thank you, Jérôme. Once again, what we are trying to do today is to give you the figures to help you better assess the potential of co-financing because some of you apparently were afraid that, let's say, the co-financing bucket would be emptied too rapidly. Once again, there is more to come than what has been achieved in more than 10 years, right? There is still a very important part to go in our pockets. This is really important to understand.
That's great. Thank you. And just a small clarification, if I may. I guess the other sort of side of the equation is sort of savings from corporate network switch-off, I guess. Again, sort of long-term theme, but when do you think these can start impacting your sort of EBITDA? And then just around sort of fixed voice interconnect, the Arcep data suggests that there's a meaningful amount of these sort of interconnect revenues which are declining quickly. Is that something that could sort of end up pressuring the wholesale revenue beyond this 2023 kind of timeline? Thank you.
Thank you. We'll now take our next question from Giovanni.
No, but we did not answer yet.
I apologize.
Ramon, maybe I can just have a quick answer because I think it's too early to answer the question, but we started to organize the hand of the corporate, and we have currently some savings, but it's too early to have a real impact because the weight of the corporate line is still very important, specifically for our competitors. We will, and I think the most important point to understand is we will manage, and we will manage the hand of the corporate with a priority to preserve the value, the value with the revenue and the value linked with the saving costs we can make. We can maybe come back on this point in 2021 because I think we will have more information, but the real point to understand is we will manage the hand of the corporate with the priority to manage the value.
Jérôme?
Okay.
Concerning the Q3, first, you have to take into account that during the crisis in France, interconnection increased because the number of calls increased, so there were more renewal of interconnection. On the long term, for sure, this interconnection will decline because that's clear. That's so clear. You see, it's a symmetrical flow of traffic. It means that we get revenues, we lose revenues, sorry, but we will also save money because we have less costs. This interconnection decline has no impact on EBITDA.
Merci. Merci, Jérôme.
Are we ready to take the next question? Okay. Our next question comes from Giovanni Montalti from UBS. Please go ahead. Your line is open.
Hello. Thank you. You were referring to the improvement in terms of pricing on mobile on the entry level for data.
I was wondering if you see ground to see some improvements starting also on the fixed side, especially considering that you're seeing so much traction on the fiber side. Finally, if I may, sorry to go back to the tower topic, but when you say control, you mean having a 50% + 1% or, in any case, being the largest shareholder of this vehicle, or would you consider other options? Thank you.
Thank you, Jérôme. Fabienne, on fixed pricing?
Yes. On the fixed, we observe the same good news, an encouraging price recovery too. We note a real improvement in pricing, especially on fiber, so it's really important. We are around for the low-end price around EUR 20. If you remember, we were around EUR 15 in Q2. We observe exactly, as for the mobile, the same movement, very consistent with the premium strategy.
It is good news. I think you can observe that in this context, Orange recorded a very strong performance on the broadband and driven by FTTH with 360,000 net adds. It is the best quarter ever recorded. Still with many acquisitions, 53% are new customers, and with a premium pricing very strong, as reflected by the ARPO. In the fixed as in the mobile, we are really confident with what happened on the French market and the repair market we can observe.
Thank you, Fabienne. On what is control? I mean, control is control, right? The important addendum to this answer is what I said about the arm's length contractual relationship with the MNOs, the dedicated management, the appropriate governance, the ad hoc reporting, etc. Looking at control, you know I think what is control.
We will go into details in February when we will have this detailed presentation with all the KPIs, which I think will help you to have also a better view on the value of these assets.
Sorry if I may have a very, very quick follow-up. Considering, let's say, how much things are improving in France, would you feel comfortable enough to expect some acceleration of the revenue momentum in terms of retail services in France?
Yes, we can. It is exactly what we announced in the past, retail services revenue improvement, and we are. Our strategy is to keep a price premium between corporate and fiber. It is exactly what we did from two years now. We are confident that we can maintain a price premium between corporate and fiber, and we can maintain a premium strategy on fiber as reflected by the convergent ARPO.
Yes, we are confident in this way.
Thank you very much.
Thank you. Our next question comes from Stéphane Beyazian from MainFirst Group. Please go ahead. Your line is open.
Thank you. Can I come back on the dividend and congratulations for fighting on this one? Was there anything sort of to give away perhaps to the French government in order to have that decision at the board? What I'm trying to guess is whether there are any conditions attached to that that could restrain your ability on the EUR 1 billion net savings plan on which you're still working on and if there's any update on this plan. Thank you.
Thank you, Stéphane. No, there is absolutely no condition, nothing to give away. It is just the acknowledgment that the resilience of the performance puts us in a condition to pay this dividend. That's it.
There is nothing more to it. Companies facing the COVID crisis are in different situations. We could make the case that Orange is facing this situation, and it is not totally surprising with the resilience you can see in these results. No conditions attached. Of course, on the cost savings exercise, the EUR 1 billion plan, we are working on it. We would like to report on this at the full-year results meeting in February. February will be a time to talk about many things. The team is working. We have a dedicated team. We're working on all the four pillars that we described. What I propose is to just wait until this time in February to go into more details, but you can trust us.
There is kind of no alternative then to fully deliver on this, looking at the smart spend initiatives, the mutualization, the digitalization. We have been learning on digitalization a lot in the lockdown context, and this is triggering many discussions on many fronts. Just to give you one figure that we were sharing with Fabienne earlier, you have now 38%—Fabienne will correct me if I'm wrong—38% of the fiber sales which have been done through digital channels. This is something which was not really totally intuitive because it was really shops and etc. Close to 40% has been made through digital channels. This is going to accelerate. Of course, with many other issues, this will be generating savings, and all the team is working hard to make sure we'll be on track with this target.
Very clear. Thank you.
Thank you.
We will now take our next question from Andrew Lee from Goldman Sachs. Please go ahead. Your line is open.
Yeah. Morning, everyone. Just two hopefully quick questions. Firstly, on your guidance for EBITDA, I just wondered, given you are down 0.6% year to date on EBITDA and presumably COVID or at least roaming headwinds are lighter in Q4, why did you not have the confidence to raise that guidance to maybe flattish? I just wondered any commentary around underlying trends within that would be helpful. Secondly, just on the mobile infrastructure, a helpful update today. I just wondered if you could talk about how you generate more actual value from this. Aside from actually spinning it out, where does the value come from? Do you think you can generate more revenue growth from these towers? Any numbers you could put around that would be helpful. Thank you.
Thank you, Andrew. Question one, why don't—I mean, given we are at - 0.6%, why don't we change something? We said around -1% . If it's around -1% , the minus is towards the minus, right? Meaning closer to the flattish point than to above the -1% . I mean, it doesn't make any sense every quarter to continue to adapt. We'll see at the end of the year, but we are extremely confident, extremely confident with this target, and extremely confident also on the organic cash flow commitment and target to deliver more than EUR 2.3 billion. I mean, you can be extremely confident with this. This is first point. On the mobile infra, if we say everything today, this is only quarterly results. You will not be there with us in February. You will be too bored because you will know everything.
You have to be a bit patient. You will have all the KPIs, all the KPIs, because you will, as I said, have a picture of an independent structure with negotiated MSAs. We're not going to give you one figure today and another figure in three months. I don't think it serves the cause. You will have the right, the definitive figures in February, and you will have all the means to assess the value which is here. When you say, "How do you generate more value?" Well, because you are managing this infrastructure as an independent business. You will look at tenancy ratios. You will look at the way you participate, for instance, to market consolidation with these assets, etc, etc. A lot of ground to cover in February, but as you can see, the work is well on track.
Thank you.
Thank you. Our next question comes from Sam McHugh from Exxon. Please go ahead.
Yeah. Morning, all. Thanks for all the detail on co-investments to be received and received already. My question was more about the co-investment fees that you have to pay yourself to either SFR or to others in the PIN areas. I wonder if you could give us the same detail on how much co-investment you have paid to date and how much you still think you have left to pay in the medium-dense and PIN areas. Thanks very much.
Ramon, you want to take the question?
We are not, I'm sorry, we are not going to give you a detailed answer to this because this is fully taken into account in our OpEx and CapEx perspective and guidance.
As you know, we as an access provider are going to fight hard to keep our market share, and we have been quite successful in doing so. We will do this accessing to third-party network, either through renting, through co-financing, through special purpose vehicles, different means. This is fully embarked in the perspectives that we have given on CapEx and on OpEx.
Thank you. We'll take our next question from Emmett Kelly from Morgan Stanley. Please go ahead.
Yes. Good morning, everybody. Thank you for taking the questions. Just two questions from my side, please. The first question is on 5G now that the auction has been complete. I guess 5G will be quite different to 4G because when I think of France, 4G was kind of launched at the same time as Free launched its mobile offering, so maybe some of the upside was launched there.
Can you maybe just say a few words on what your plans are for 5G and whether you can maybe go early on 5G, go big on 5G, and maybe take some market share, maybe a little bit like what Deutsche is doing in Germany? Just a second question. Maybe you probably won't be able to comment on this, but could you say a few words about the Sogicom tax issue with the French government? I saw some media articles in France last week suggesting a potentially positive outcome for Orange there. Thank you.
Okay. Thank you, Emmett. Maybe on 5G, Fabienne?
Yes. You're right. The context to roll out 5G is very different than when we launched the 4G in France. We are in a better context, and this is the most important point.
As I explained just a few minutes ago, we have a repair market on the mobile, and we observe that all competitors announce their price with the 5G. In this context, our strategy is still the same. First of all, we want to maintain our leadership on the mobile network and the premium we have on the network. It's very important for Orange in all territories and in all areas. This is the first point. We will start to deploy 5G tomorrow because the auction has just closed a few days ago. We want to maintain our leadership. In this context, we want to maintain our premium strategy. As announced with the launch of the new plan we made in October, if you remember. This new plan is not only 5G, but is 5G ready and embedded a premium pricing as announced.
No difference for us, the wish to be leader and to maintain a differentiation on the mobile and in all territories and in this case, to maintain the premium price. We are ready to launch now, and we start tomorrow.
Thank you. Thank you, Fabienne. On the second question on the tax issue, it's an issue which has been going on for close to 10 years. The good news is that we will have the final decision of the highest French court, Conseil d'État, before the end of 2020. It brings bad luck to talk too much about expectations about what could be good news. We will be waiting for this decision, and voilà. Merci.
Thank you.
Thank you. We'll now take our next question from Thomas Coudry from Bryan Garnier. Please go ahead. Your line is open.
Yes. Thank you. Good morning, everybody.
Just a follow-up question first on 5G. What are, beyond, I would say, the market, the improving market environment, what are your convictions about the willingness of customers to pay for premium in 5G? We saw that in some countries, there was some disappointment about the technology. My question is really, how much you believe that a premium is sustainable there? My second question is actually on OBS France. Iliad should launch its activity in B2B soon. My question is, how did you prepare for that? Is it a non-event for OBS France? Did you take any specific actions? It seems that most players on that market are kind of downplaying the entry of Iliad. I would like to have your point of view on that. Thank you very much.
Okay. Thank you, Thomas.
I think you are the last question, so we will close following the answers to this question, and Fabienne will take this. Thank you.
Yes. Thank you for the question. It's complicated to answer your question for the future, but I can give you some figure now, and it's very important to know two points. 65% of French people are interested by 5G. Okay? This is the last survey we have. Despite a lot of debate in France, 65% of French people are interested by 5G and are ready to pay for 5G. Okay? Currently, we have an experimentation in France with 600 towers in 5G and around 1,200 people in experimentation in 5G. For all these people, we gain a better NPS with a gap around 20 points.
This population, these people who are tested, who are in experimentation in 5G, are ready to pay more for more. It's exactly the strategy we defend from a few years now. We are confident, and it's exactly what I explained just two minutes ago. The new plan we launched in October is really successful for the moment. The beginning of this new plan is very good because we give more for more, not only 5G. We give more data. We give more SIM cards. Okay? You're right. In some countries, there is deception recorded because in some countries, they decide to launch 5G with two technologies: the spectrum dedicated to the 5G and the DSS. It's a dynamic spectrum. The quality on the DSS is not the same as with the spectrum dedicated for 5G.
For Orange, we will be very cautious about the quality of the network we will roll out, and we want to give and to maintain a premium on the network and a high level of quality of our network. This deception is only in the country where you have a mix between these two technologies. For the customer, there is no difference between 4G and 5G. We will be very cautious about this point. I am really confident that 5G will find the attempt of the consumer. It is exactly what we can prove currently with the experimentation we have.
Thank you.
Thank you. We have no further questions. I'll turn the call back to the speakers for any additional or closing remarks.
Thank you. Thank you very much. Thanks to all.
I would just maybe like to conclude with maybe three, four very, very quick points. First, you see this quarter that the operational performance, the commercial performance is very good. It's much better in some critical countries like France and Spain, but it's very good everywhere. In the other European countries, in Africa and Middle East, we did not have any question on Africa, but you could see how dynamic the evolution is there. Second, the EBITDA trend is healthy because it's fueled by this performance. This is good. Third, the co-financing is still going to be with us and to fuel EBITDA with more than 50% still ahead. Fourth, the cash flow, as you can see and you will see, is well oriented. This explains the positive move we were able to announce on the dividend. It's a good quarter. It's a very difficult period for everybody.
We'll be working hard and hoping to see you soon and hoping that everybody will be well in this complicated time. Thank you very much.
Thank you. Ladies and gentlemen, this concludes today's call. Thank you for your participation. You may now disconnect.