Good afternoon, ladies and gentlemen, and welcome to the Proximus Q2 2019 Results Conference Call. For your information, this conference is being recorded. At this time, I would like to turn the call to Madame Nancy Goossens, Director, Group Investor Relations. Madame, please go ahead.
Thank you. Good afternoon, ladies and gentlemen... And that you have been able to go through the results. We keep our usual format for this call, meaning that we will use most of the time to answer your questions. I have here with me the CEO, Dominique Leroy, and CFO, Sandrine Dufour, as well as other members of the executive committee. They will all be happy to take your questions in a minute. Before we get to that, we will first listen to the introduction of Dominique. Please go ahead.
Thank you, Nancy. Hello, everybody, and welcome to our Second Quarter Conference Call. As you could read in our announcement of this morning, we managed to further grow our customer base in the second quarter, strengthened by our dual brand strategy, and this in a market which remains highly competitive. We further attracted new customers to our main products, growing our customer base both for fixed internet and TV, and this for both Proximus and Scarlet. On the mobile side, the Proximus consumer segment returned its growth in mobile postpaid back to normal, increasing its mobile postpaid customer base by 19,000 customers. More and more of our customers take both fixed and mobile subscription. Hence, our conversion customer base progressed further.
This was especially driven by enticing families to our all-in offers, Tuttimus and Bizz All-in, but also on our Minimus offer and the Epic Combo offer, which is more oriented towards millennials. As a result, the average revenue per household showed a further slight progress and grew to EUR 66.8. On the enterprise side, we also achieved to further grow our mobile customers on an already high base. Competition is, however, increasing in this domain, which is reflected in the net mobile customer growth and ARPU. On the fixed side, we see, however, encouraging support from our point-to-point fiber part, benefiting the new data connectivity solutions, and therefore softening the impact of the ongoing erosion of legacy services. On the ICT front, we continued to post a positive revenue variance, including the support of acquired high specialized companies.
Revenue from ICT is typically more volatile, more precisely for one-shot products and services needs. We posted higher revenue too in advanced business services, essentially for our subsidiary, Be-Mobile, active in the field of smart mobility and benefiting from its acquisition of Mediamobile. With positive commercial drivers in both the consumer and enterprise segments and revenue pressure partly on low-margin income, we kept a sound underlying domestic direct margin. Remember that in Q2 2018, we benefited from significant tailwinds. These aside, we would see our direct margin being slightly positive year-over-year. This in spite of regulatory impacts with lower fixed-termination rates, and since mid-May, also lowered international calling rates impacting our margin.
As we further progress on our digital journey, we are able to materialize cost efficiencies in our domestic operations, which have helped to achieve a stable cost level despite the higher personal costs related to ICT. For the second quarter, this has led to our domestic EBITDA to be 0.8% lower compared to a high base in 2018. For BICS, the international carrier segment, the trend of moving from voice to data continued, and we saw again a strong support from TeleSign, showing a solid increase in the application-to-person messages. The direct margin of BICS was up year-on-year, yet more than offset by some higher OPEX costs, amongst others, to support the current and future growth of TeleSign. Based on the achievements so far, we reiterate our full-year guidance.
For our domestic revenue excluding terminals, we expect the next half of the year evolution to be in line or slightly better than the first half, depending on the more volatile part of the ICT revenue. The group EBITDA is expected to be stable with, within the mix, a slight increase of our domestic EBITDA, offset by a lower EBITDA for BICS. So far, the announced renewed agreement with MTN, which foresees the progressively insourcing by MTN of their Africa and Middle East operations, did not have a significant impact on BICS. We expect a more pronounced impact as from Q3. Our CapEx outlook of the year remains unchanged as well. We expect to end 2019 with CapEx stable to the previous year, excluding spectrum-related CapEx.
On a different topic, as you can imagine, the next months we will work hard to further develop the outlined mobile access network sharing agreement with Orange Belgium. We were pleased with the 11 July announcement, seeing the numerous benefits it will bring for our customers and for us as a company. As a final point on the regulator consultation on cable wholesale tariffs issued early July, I have to say, we were surprised by the low cable wholesale pricing. We are now awaiting the consultation on the fiber pricing, which we expect in the course of September. As we have clearly stated before, we consider a fair regulation, which means that the conditions to invest remain appropriate as a key for any future decision on our fiber deployment. We will give more news on that with the full year results of 2019.
With this, I have covered my introduction and propose we now start with your questions. Thank you.
Ladies and gentlemen, we will begin with the question and answer session. If you wish to ask a question, you may press zero one on your telephone keypad.
Operator, can we start the questions, please?
Ladies and gentlemen, if you wish to ask a question, you may press zero one on your telephone keypad. We have one first question from Mr. Michael Bishop from Goldman Sachs. Sir, please go ahead.
Yes, thanks, and good afternoon. Just two questions from me, please. Firstly, you mentioned that you were quite surprised by the low cable wholesale pricing. Do you think that provides any read across in terms of the FTTH pricing, or do you think the direction could be independent? Maybe just a quick follow-up on that, how much do you think that, really, even if the FTTH pricing is higher, the fact that the cable pricing has now been cut or proposed to be cut by so much, that provides a disincentive to invest from your side? Then secondly, I was just keen to get any early signs on how you're seeing the uptake of the Epic Combo deals, particularly with relation to Orange Belgium also announcing its standalone broadband offers recently. Thanks.
Concerning your first question on the cable wholesale price versus fiber to the home, the regulator in Belgium and already in this market analysis, has come with the hypothesis that there were two markets in Belgium. One market which was the fiber and copper market, and another which was the coax market. I think they will do their cost model a bit differently on the two sides. On the coax, the fact they come with a low price is very much linked to the fact that everything which has been amortized is not taken into account in the cost model, which was a surprise for us and will of course be the comments we will make on the consultation period, which is open now till September 6th, is what we see here.
On the coax, it's mainly amortized costs which are not taken into account, and this will be mainly the comments we will give. I don't think there will be a read across on the fiber, because on fiber there is not a lot that has been amortized. In that sense, we guess that the wholesale price from the regulator for fiber will be higher, and more in line with what we currently have as commercial price. Of course, the issue is that the market is not two markets and that fiber and coax are one market. The read across you can do for the fiber investment is most probably that will be more complicated for us to attract wholesale customers on our fiber network because the coax will be sensibly cheaper than the fiber.
That's why we are now currently waiting indeed to have the final fiber prices. We will also wait to have the final coax price. We will then come back to the market with our fiber deployment plan, which will most probably be together with the year-end results that we will bring February 2020. For the other side, on the fiber, we are currently continuing the deployment. We are currently in 11 cities in the country. Deployment is proceeding according to plan. Win-back and migration is also proceeding according to plan. I think from a business perspective, fiber deployment and fiber take-up are in line with our plan. I think the main uncertainty, which is still open, for the future acceleration or not of fiber, is mainly linked to regulation. I'll let Guillaume answer you on the Epic Combo question.
On the Epic Combo question. First, just to renew of the strategies that we are trying to deploy, we want to drive conversions through our segmented approach, with a multi-brand strategy. What we did in the previous quarters, and before the announcement of Orange, is that we did two main things. First, for the pure price seekers, we launched a new 1P Scarlet offers, to reinforce the leadership of Scarlet for the price seekers. We launched that in early end of May this year. Prior to that, we indeed launched our new Epic Combo offer for the higher end of the market. Also for a segment of millennials that are looking for worry-free mobile data bundles and internet access. A double play mobile plus broadband.
It's way too soon to say how everything is performing as we speak because the Orange launch is very recent. What we can say, that we are growing on every segment of our customer base, our Scarlet, Proximus and Epic. Last but not least, the way we designed Epic Combo, we think it's much better proposition for this segment. Why is that? Because on top of mobile and broadband access, we also provide OTT TV, which is very important for this segment of the market, including, of course, live streaming of television and broadcasters, television channels on all devices. We think that we are very well prepared, that we anticipated very well the competitive move of Orange. We are quite confident that we'll be able to continue growing on every brand till the end of the year.
Thanks. Can I get info from that? Basically, you're not really seeing a big increase in the proportion of gross adds coming through the Scarlet brand. I think in the past you'd said it was it around 10%?
Sorry, could you repeat the question because it was a little bit noisy here.
Yes, sir. I was just picking up on the Scarlet point. I'm just asking if we can infer from that you're not seeing a big increase in the proportion of gross adds coming from the Scarlet brand. I think from memory, you'd said in the past that it was a fairly low proportion, around 10% or so.
We do not see any major evolution of the share of Scarlet's gross adds within the total mix of Proximus. That's correct.
Okay, great. Thanks a lot.
Thank you, sir. We have another question from Mr. Nicolas Cote-Colisson from HSBC. Sir, please go ahead.
Hi. Thank you. Can I come back on what I'm reading as kind of a weak mobile net adds in the quarter? Especially for EBU, considering that you had, if I remember well, a big contract win in Q1, which was supposed to have some spillover effects in Q2. If you can give us a bit more color on the trends, and who do you think is the most active competitor, both in business but also in the consumer segment? A similar question regarding ICT. I hear what you said, Dominique, about the volatility of the business there. Given there is still some effect of acquisitions in Q2, it seems that there's very little organic growth left here for the quarter. If you can give us more color about longer term trends here. Thank you.
Okay. This is Bart speaking. I hope you can hear me because we have here some problems...
Yes, we hear you.
... Conference call system. It is indeed so that in terms of mobile adds, we still grow in this quarter with about 5,000 net adds. It is also true that we had a large contract with the Flemish government, of which the most of the cards have been activated in Q1, but still an important part has been activated in Q2. You're right in the sense that Dominique already indicated in her introduction that competition is increasing on mobile in enterprise, that we don't keep growing in the same pace as in the past quarters is not such a surprise given the high market shares that we have in this environment. That is for mobile. On your question on ICT. Indeed, ICT is a more volatile market. You're right. From one side, we still have the impact of the M&A of Codit and Umbrio mainly.
What you see is that the M&A impact is not fully, how say, leveraged in the ICT results because next to that, there is lower results in the core ICT markets. What is important in that sense is that it's mainly products, low margin products and one-shot services. I think what is important is that we keep growing in the higher value services, especially in those affiliates, so that we get to a better mix in terms of profitability, and that is completely reflected in the direct margin evolution, where we see a direct margin evolution of plus more than 3% direct margin growth. Just given we have those higher margin services and a better mix in the portfolio.
Bart, if I may, just to follow up on your first point on EBU and mobile. You're mentioning that there's a fight for subscribers in the market. How does it go in terms of pricing negotiation? Can you give us a feel of what type of discount you have to offer to companies in order to win big contracts? Obviously, a range, not giving secrets away.
I didn't get your full question, but I think your question was, one, who is the main competitor? Two, what are the discounts that we have to offer to keep winning?
Yeah.
The main competitor of mobile in the enterprise market is certainly Orange, but also Telenet is active, but its main competitor is Orange. In terms of the discounts, well, I can only say you have seen that our ARPU had a decline in mobile, and that decline is from one side linked to the competitive pressure, from the other side linked to the move to b undles where we have less out of bundle. Those are the two impacts that we have in terms of pricing.
Okay. That's very clear. Thank you very much.
Thank you, sir. We have one question from Mr. David Vagman from ING. Sir, go ahead.
Yes. Thanks. Good afternoon, everyone. Thanks for taking my question. If we can come back on regulation and, let's say, your interpretation of the wholesale cable rates, the pricing and the cost model. Let's say that these models and pricing are confirmed, should we understand that you would be ready to slow down your fiber rollouts because of regulation? What kind of conclusion should we draw? Secondly, on consumer mobile and on the commercial momentum, could you help us understand the dynamics of the Q2 performance? Let's say distinguish between Scarlet, on one hand, Epic Combo, the normal Proximus offer also. Finally, let's say the collapse of the Mobile Vikings client base. I think they lost something like 100,000 customer in Q2. Did you see a flow of Mobile Vikings clients? Thank you.
Coming back on the regulation, I didn't say we would slow down the fiber rollout. I think what we have said very often to the market is that we were considering to accelerate potentially the deployment of fiber. That's something we would consider based on the deployment costs, the business case, win-back and migration upsell, and the regulation. I think what we see now is that on the parameter we can control, I think we are in line with our business case. If we look at regulation, it doesn't help the fiber case to have a coax regulation which is extremely low in terms of price. On the fiber, of course, we have always planned to roll out our own fiber, but we were also looking at having wholesale customers on our fiber.
Today you know that Orange is on the coax, which is the biggest wholesale in the fiber to the home business. I think on the fiber to the enterprise, I think our fiber and our copper are still very much answering the needs of our own customer but also on wholesale customers. The debate is currently mainly on how will we be able to attract potential wholesale customers on our fiber deployment for the fiber to the home. That's something, if prices stay on the coax as low as it is, it will be probably more complicated. In that sense, we will have to redo the whole math, which was foreseen by the end of the year, and we will come back to you with more information by then.
I think so far we continue the plan as it is, and we will come with an update by February next year with our full-year results to give more color on our further fiber rollout. At this stage, nothing has been changed to the current deployment plan of the fiber.
On the dynamics of the Q2, you know that we are not disclosing precise numbers on the evolution of every brand and sub-brands. What I can say is that we are growing both on the Proximus brands and including Epic Combo and on the Scarlet brand. This is true for broadband, and it's obviously also true for mobile. On broadband, I think we were quite happy with the performance of Q2, thanks to good acquisition but also a stable churn on both products. We achieved this without being very aggressive in terms of promotion. Despite a very intense competitive environment, for us, it's been a rather reasonable promotions activity. We are quite comfortable with also the remaining part of the year where we think that we can also play a good position on this very competitive environment.
Okay. Thank you. Maybe on Medialaan and Mobile Vikings, did you see any impact?
On Mobile Vikings, the impacts are rather minor as we speak. There is no massive impact of the launch of Mobile Vikings on the market. I would say it's too early maybe to see what could be the traction for this kind of offer. So far we see no impact neither on Scarlet nor on Proximus.
Okay. Thank you.
Thank you, sir. We have next question from Mr. Emmanuel Carlier from Kempen. Sir, please go ahead.
Yes. Hi, good afternoon. Couple of questions. First of all, on the subscriber additions. I think it's fair to say that the trends remained quite soft. If I remember well in the Q1 call, you guided that momentum would improve in the coming quarters, partly driven by more aggressive marketing from your end. Is that still something that you expect? If not, why not? What has changed? That's the first question. The second question is on the launch of Orange Belgium's Love Duo offer. I think it's fair to say that broadband on a standalone basis is still not that popular in Belgium. Could you share your view on how important this market Might get in a couple of years' time. How you believe you can avoid, I would say, unbundling and also avoid value getting lost here. Thank you.
As I said, on net additions, as you said, we think Q1 traditionally for Proximus is not the best period of the year because we have the churn rate of the big volumes of the Christmas market. As foreseen, we improved significantly the performance on mobile with a little bit more marketing activities, but without being excessive in that domain. We managed to go back to the normal pace of growth of Proximus on mobile without being very aggressive in terms of promotion, which is important to note. The second part of the year, of course, is back to school. You have the Black Friday and then the Christmas Market. In terms of promotional activities, it's always for Proximus a much intense period. We do think that it's going to be the same this year.
It's nothing different from what we did in the previous year. For sure it's going to be more intense in terms of marketing means and commercial activities on both brands, both Scarlet and Proximus. On your question about the size of the 1P market, for sure it's not going to be more than 20% in terms of maximum market size for a 1P only offer in the medium term. For us, what is very important is to work on the positioning of the brand, on the relevance of the brand. This is why we did announce major innovations the 13th of June, repositioning of the brand with the Think Possible tagline, and also a lot of new services around gaming, around digital services that could help in the future, help the Proximus brand and all of the brands being relevant for our customers.
Doing so, we should sustain the value creation for our customer base. This is the way we want to differentiate ourselves, being innovative, launching relevant digital services, including gaming, and pursue brand differentiations around conversions and digital services. Of course, for the price seekers and those people that want to really have a 1P offer, we do have today the best offer on the market with a 1P broadband offer at EUR 32 per month, which is really attractive, and that is today with Scarlet meeting its audience.
Yeah. Thank you. If I look at unbundling, is that something that you expect in the coming years? If I look at what clients want, they basically just want fixed broadband, plus mobile. Maybe they want a bit of TV content, but they don't want to pay a very high price for that. That is the main risk, I would say, to the sector. With the cable regulation, it's a bigger threat in Belgium probably than in other markets. Could you share some thoughts on, not next quarter, but really on the three to five-year view, how you think about this topic?
I will try to answer a bit on this question. I don't think the Belgium market is so different from other markets. I think Belgium has been a market which has been a market with a lot of convergence quite early. We see that trend is still there. What you see on convergence is that even on convergence, you seek to start some more segments. You have people that really want the full product with the 4P and the content where they can find that in Tuttimus. We see indeed more people now wanting to have a triple-play mobile fixed offer with the mobile internet and TV. That's what we currently offer on the Minimus proposal. You see more millennials who are more indeed in the mobile fixed internet and with an OTT proposal, which we offer on the Epic Combo.
The broadband-only segment is a segment that has always been available, both under Scarlet and Proximus. Scarlet has broadband at EUR 23 and EUR 32 with the unlimited offer at EUR 32. We see that that offer is increasing, but it's not increasing more today than it was over the last few years. I don't think that you will see a lot of unbundling. I think unbundling will be there for a small part of the population in Belgium. Millennials will take unbundling, but not pure broadband only. They want broadband and mobile and OTT content, and I think that's what we offer. If you look more at families, they want an offer with more content because they have kids, they have teens, they want gaming, they want content.
I understand your questions and I understand it comes from probably some talks that have been recently held by Orange on the launch of their proposition. I am not convinced that there is such a big market for broadband only. I think the market is segmented according to the various segment of the population. I think with our portfolio, we are probably the only player in the country that is able to really offer the right solution for each segment. I think that's what makes us way more resilient than some others to all the movements in the market. That's what make us confident that we can still continue to grow our customer base and drive value in the Belgian market.
Okay. Thank you.
Thank you, sir. We have another question from Mr. Roshan Ranjit from Deutsche Bank. Please go ahead.
Great. Afternoon. Thank you for the questions. Two for me, please. On the mobile side, just referring to your network sharing announcement with Orange Belgium. Can I just check on the open tower regulated structure, does that apply to both macro sites and rooftop sites? Therefore, if so, will the incremental sites that you are looking to build, I think around 800, will the open tower regulation apply to those? Part of that question, how are the discussions going? I think Telenet was looking to get involved with your partnership with Orange Belgium and yourself. Secondly, Dominique, you mentioned the scope for maybe accelerating the fiber rollout. Given your agreement with Orange Belgium on the mobile side, and I think there's still a keenness to potentially co-invest, is there still an opportunity for that to happen and therefore accelerate your fiber rollout? Thank you.
Okay. I'll try to answer all your questions. Concerning the mobile network sharing and the tower regulation, it's indeed true that in Belgium you have a passive sharing obligation, which is both for the macro antenna, but also for the rooftop antenna. That's something which is there and which will apply to all the sites. That's for big and small sites. Concerning the request and the demand, indeed, from Telenet to join, I think that's something we are open to, same as Orange. I think what we will need to consider in that is that, is there a technical feasibility to host on the new equipment for the 2G, 3G, 4G and tomorrow 5G, three operator on one site, which is not obvious. Of course we will look into that. So far, the discussions with Telenet have not really started. There is holiday periods.
We are looking at all the NDA process because all this network sharing between competitors needs to be taken care of extremely carefully because there is a lot of sensitive information. It's all about setting the right process and the right structure. Today we are more looking at how to organize ourselves to have Telenet trying to join the party, and then we will look how to do it, what is the feasibility, and if indeed it makes also economical sense. We are open to look at that. On the fact that we are now a bit closer to Orange, thanks to the network sharing, that's true, but we are still strong competitors.
While we found a deal on the mobile network sharing, because we think there we can really create value first for the customers, because we will have better coverage, better indoor coverage, and faster rollout of 5G. The main goal is to bring indeed the best network at the lowest cost, but as fast as possible, also 5G. We remain strong competitors, and we remain strong competitors in the mobile, but we also remain strong competitors in the future fixed. I think there, we of course would like to co-invest with Obel or having Obel on our fiber network, but that will be a discussion based on economical reality and also on the wholesale pricing, which we are still waiting for. I refer to some of my answers of previous questions.
Okay, great. That's very helpful. Thank you.
Thank you, sir. Next question from Mr. Ulrich Rathe from Jefferies. Sir, please go ahead.
Yeah, thanks very much. I wanted to address the question of customer intake, in particular with regards to the split between the Xplay. It looks as if the quad-play intake slowed down further from the first quarter. I'm just wondering whether that's an element of saturation in the market or whether that's just the way you direct your marketing efforts at the moment, or whether that's indeed a competitive issue that you think could be transitory. That'd be my first question. The second one is, you highlighted the underlying growth in the gross profit. I think in the domestic business, it grew when you exclude the one-off, I think, on your numbers. That's obviously a key point when the revenue declines.
Do you think that sort of stable gross profit can be carried over into the second half of the year and then maybe even into 2020, if you're already willing to discuss at that level? Or, is the revenue decline at some point going to affect gross profit and really put the cost-cutting measures in OpEx under pressure? Thank you.
On the first question with the slowdown of 4P on that quarter, I would say it's a bit too early to speak of real trends on this one. We do see indeed a nice take-up on convergent 3P. Convergence mobile plus a 3P offer. What is very important for us on the longer term is to stimulate convergence, being through 3P mobile or 4P. We're going to drive our strategy to grow the average revenue per home. As long as the average revenue per home is growing, I think that we are doing a good job on the customer value-based management.
This is what we're going to do in the future, focusing on the value of our customer base and focusing on convergence being 4P or 3P. The main obvious reason for that being that a convergent customer home being 4P or 3P has a way lower churn than the other customers.
On your second question, I think it's a bit early to discuss the 2020 trends. If I focus on the second half of the year, well indeed we had a bit of a soft revenue, at least this is consensus. As you saw, we were able to cope quite well at the level of direct margin. This is also linked to the fact that part of the revenue which is under pressure comes with low margin between the ICT or we discussed as well some trends in the mobile with inbound revenues, which is decreasing and have no impact on our margin.
When we look at the second half of the year, we aim to continue to have a very good cost control as we did deliver in H1, so that we are in a position to deliver on our guidance of a slight growth at the level of the domestic EBITDA and stable at the level of the global group EBITDA.
Thanks. If I may follow up on that answer. I think there's a difference between having to offset declining gross profit with OpEx measures because that's a sort of a treadmill in telecoms. It's a different thing if you essentially drive efficiencies and you have more or less stable gross margin. I was particularly interested whether you were going to give us any indication on the gross margin or the gross profit rather in the second half of the year then.
Well, if you exclude the one-off, it shouldn't be very different from H1. That's in terms of expectation for the second half.
Thank you.
Thank you, sir. We have another question from Mr. Alexandre Roncier from Exane. Please go ahead.
Hi. Thanks for taking the question. I was just wondering if you could update us on the current negotiation with the unions, if you were seeing progress here or when were the next date we could see any announcement, perhaps. Thanks.
Concerning the discussion with the unions, we started beginning of January with information and consultation phase. We had a lot of explanation, discussion, dialogue on the various plans. As from beginning of June. The 10th of June, we started in a negotiation phase. You know, due to the holiday, we were not able to go very far in the negotiation also because we also had the network sharing, which was also a moment where we had some discussions with unions. We are now still continuing from all sides, preparing all necessary documents and details plan so that we can start back with the union negotiation on hopefully a much faster speed as from 20th of August.
Our objective is then to end the negotiation certainly before the end of the year, so that we can hopefully have already some of the benefits of the Fit for Purpose Plan for next year. What you see in this year cost control is that we still have some ways of the former Early Leave Plan, which still impacted 2019, and we will still have some departure beginning of 2020. We, of course, this year, in the cost control, we are also not hiring as many employee as we used to do.
If we have natural leaves and natural attritions, we tend to cope with it on the short term to be able to then, through the fit for purpose plan, train and reskill people and fill then the vacancy as we go so that we can then have potentially a lower impact on dismissal and a lower number of hirings by trying to reskill, retrain our own employees into the vacancies we have and we will have after the next quarter. I hope that gives a bit more Clearness and clarity, sorry, on where we are. We will resume the negotiation 20 of August, and we hope from then to be able to go fast, because we will already have a lot had been discussed, and a lot will be ready in terms of detailed documents.
That's very helpful. Thank you.
Thank you, sir. We have next question from Mr. Paul Sidney from Credit Suisse. Sir, please go ahead.
Thank you very much, and good afternoon. Just two questions from me, please. Just the first one, on the balance between customer acquisition and profitability. You obviously lost a little bit of traction in H1-19, and I know some of that was Proximus' own decision, but how should we think about the balance between profitability and your desire to maintain EBITDA stable or even grow EBITDA in the future, and balancing that with subscriber growth in an increasingly saturated market? Just secondly, coming back to the cable wholesale debate and your own FTTP ambitions. If the FTTP rates are significantly lower than you originally planned and the rates that were proposed in December, to what extent can you resell cable from a practical point of view, from a regulatory and a political point of view?
I know we'll see the June 18 ruling. If you don't build any of your next generation network, does that mean you can potentially sell cable across most of Belgium? Just wondering how to think about that. Thank you.
I'm not sure I fully understood your first questions, I think for me, it's an and-and. We need to continue to have strong customer acquisition, I think everything we do, both on developing new solutions for the enterprise business and building a strong brand platform, Think Possible and Pickx, and the whole segmentation offer at the CBU should help us and drive us into further customer acquisition. At the same time, like most of our colleagues, we are really trying to see where can we generate more profitability, more efficiency, drive our costs down by being more efficient, doing things first time right, doing more digitalization, doing more automation, doing sharing deals as we have done one with Orange on the mobile. For me, I would not oppose one versus the other.
I think we should continue to get good customer acquisition, and we should continue to get cost down and drive profitability as we have been doing over the last years. On your second question, so concerning the access of Proximus to Coax. Unfortunately, we do not have free hands there. We have fought a lot to be able to have access to Coax in the past after a very long legal battles and debates, which lasted several years. We have now been granted what we call a reasonable access to Coax. What is currently defined as reasonable access is that we would be able to get access to Coax in zones where we are not able to provide 30 Mb per second network speed. So far, with the quality of the network we have, it's not a lot of Belgium.
I think it's currently estimated between 5%-10% of the population. The possibility for us to get onto cable and resell our proposition on cable is there, but it's relatively limited. That's of course, also something we will need to see as we evolve and as the needs of the customer evolve. Is there anything we can do about that limit of 30 Mb per second? So far it is quite restrictive. We will do that. We have said that we would be interested in having access to cable, but on the short term, it's a relatively small part of the Belgian population.
It's very clear. Thank you.
Thank you, sir. Next question from Mr. Stefaan Genoe from Degroof Petercam. Sir, go ahead.
Yes. Thank you. Two follow-up questions. On the trends we see in the market today with lower television net adds, the launch of the Epic Combo product, and the trends we're seeing in the market today, how do you believe this mix, or what do you anticipate this mix to impact your direct margin, let's say for the coming two to three years, purely from this point of view? Secondly, on the union negotiations, did the network sharing deal have an impact on these negotiations? Have you, I would say, reviewed your initial anticipations of savings from the union deal versus the start of the negotiations? Thank you.
Firstly, on the lower TV net adds. What we see on the market that indeed in Q2, we did a little bit less TV net adds than broadband net adds. It means that we sold less set-top boxes on the market. It does not mean that we are not selling less access to OTT TV and OTT platforms. What we foresee for the coming quarters and months at the...
Sometimes we're going to sell TV access without adding a box to rent or a set-top box to give to our customers, due to the new ways of consuming television services. It doesn't mean that it's going to be less value for us, maybe less hardware, but not less value for us in the coming two, three years. Especially if we are quite good in executing our fixed platform strategy, which means to provide relevant traditional TV, but also innovative services around a Proximus platform, around content and entertainment. I don't see this having an impact on the EBITDA margin. On the contrary, it could also have an impact on the cost to serve and to serve customers, because we're going to have less hardware to provide to the homes of our customers.
On your question regarding the network sharing deal and the impact on our cost plan. First, the cost ambition we gave was for the years 2019 to 2021. EUR 240 million gross OPEX reduction is a program for the next three years, including 2019, while the ambition that we have given as part of the network sharing deal in terms of global CapEx and OPEX reduction, was a 10-year ambition. The beginning of the savings will not be visible within this three-year span, because it takes time to put the enablement and to create the common grid. That, we do not change, as of now, the program of our reduction cost for the next three years.
Okay. Thank you.
Thank you, sir. We have another question from Miss Nawal Kissi. Madam, please go ahead.
Thank you very much. I have three questions, please. Two quick follow-ups on network sharing. To start with, in your announcement, you have not mentioned that this network JV is subject to regulatory approval. It would be helpful to hear your views in terms of any possible intervention from the antitrust authorities. That's the first question. Secondly, some operators elsewhere in Europe have been excluding high-density urban areas from their network sharing, and they've been citing that it's because of data traffic management issues.
It would be helpful to hear your thoughts on this particular issue. Lastly, we've been talking about possible changes of ownership at VOO for many years now, but it seems that things are moving on that front. If that change of ownership were to crystallize, how would you think about protecting the business? In particular, would this be of any relevance to your fiber plan? Thank you very much.
On your first question, the network sharing and regulatory approval. The way the JV will be set up is first services that will operate and maintain the network. The assets will remain with the parents. In that sense, there is no ex-ante approval. In that sense, we can still have questions from the regulator. We can have questions from the Competition Authority. That's something we could not anticipate. That's why we will proceed with the JV. We will continue, of course, to be available for all questions of authorities. There is no possibility for us to ask for approval before setting up the JV. It's indeed a potential risk.
We think today in the current setup, with the size of Belgium, with the competition that is still there, with the fact that we are just doing a MORAN network sharing and we keep the core separated, the spectrum separated, and all the offers we bring in the market will remain very competitive offers from both sides of the table. We don't think there will be a major issue on the JV. I think, yeah, excluding parts of the high density or the spectrum, we have looked at different opportunities, but I think Belgium is a very small territory. As soon as you start to exclude part of the territory, it's extremely difficult to define what would you exclude. Also the fact that we want to go faster and deploy faster 5G, which is one of the main driver of creation of the JV.
It's also very difficult for us to exclude any other frequency bands because you know today, JV is a standalone network, and you need to have the integration of 3G, 4G, and 5G to be able to build the right 5G. That's why we have proposed, and we have agreed on a network sharing, which is for the full country and all frequency bands, but of course limited to the active network equipment.
Concerning. Yeah, I also read the newspaper as you most probably do, and we read indeed that VOO is, that Nethys is preparing itself in a very separated entity, which has been done recently over the summer. There are discussions with Brutélé to finalize a deal to integrate fully the Brussels cable business into this new company.
We hear as well that there would be discussions to sell the entity even potentially to a private equity, which is also mentioned next to the obvious other Belgian player. Would the sell of VOO impact our plan? I would say no. I think today we have our own strategy of deployment of fiber. We have our strategy of deployment of 5G together with Orange on the network sharing. I think in that sense, a sale of VOO would not impact the strategy of Proximus and would most probably not impact short-term or business significantly.
Very clear. Thank you very much.
Thank you, madam. We have one last question from Mr. Nicholas Prys- Owen from UBS. Sir, please go ahead.
Good afternoon. Thank you very much for taking my questions. I just had a couple of follow-ups on the earlier question around the ICT business. Firstly, just on the performance in Q2, I wondered if you could help us disaggregate the impact between the weakness in the low-margin products this quarter versus the impact of the two acquisitions you acquired in 2018. Just on, I suppose you've chosen to highlight the volatility in the ICT business a bit more this quarter. I suppose it wasn't really mentioned in Q1. I suppose I wanted to clarify how much, if any, of the weakness in the low margin revenue was a timing effect. Do you expect to support the Q3 ICT growth?
Just finally, given your comments around the revenue guidance and the volatility in ICT, I wondered if you could perhaps give us some more color around what portion of the ICT revenue base you have good visibility on for H2, and what portion of the revenue base is accounted for by revenue streams you think of as volatile. Thank you very much.
I'll take your first question. On the impact of the low-margin ICT business. When you look into our figures, you see that ICT is still growing overall. Now, the impact of the M&A, which has been highlighted, I think, last quarter by Sandrine also, it is still relatively small figures in the total. I think you should take that into consideration. Nevertheless, we are still growing. It means the impact of the low-margin product deals is there, but it is not huge. Of course, especially not in the margin. You should also take into consideration that we had in Q2 of 2019, relatively high product revenues also in ICT. Q3 impact, I think I leave the question together with the revenue guidance to Sandrine.
On the guidance for revenue, I don't think we can be much more precise at what we said. Certainly, we look at H2. We expect in H2 about the same level of performance that we had in H1. That could be improved if what we see in the pipe of ICT is being delivered. Definitely we have better pipe in H2 than the one we had in H1. Now, there are elements of timing which can play. I cannot really venture on a number, but it can be between around EUR 10 million of revenue that can be in the next H2 or slip over the next year. Just to give you a sense of what we're talking about for the second half of the year. That gives you a bit more understanding as to why we phrased the revenue guidance as we phrased it.
Again, I want to insist on the fact that the impact on our profitability will be minimal, and I think this is important that we are comfortable on our level of direct margin and level of EBITDA guidance.
Thank you very much. That's very clear.
Thank you, sir. We have no other question. I give you back the floor for the conclusion.
Thank you all for participating to this call.