Telecom Italia S.p.A. (TIAJF) Q3 2021 Earnings Call Transcript
Published at 2021-10-29 11:35:23
Ladies and gentlemen, good afternoon, and welcome to Telecom Italia Q3 2021 Results Conference Call. Ms. Carola Bardelli, Head of Investor Relations, will introduce the event. Ms. Bardelli, please.
Ladies and gentlemen, good afternoon. A very warm welcome to our Q3 2021 results. I'm here with our CEO, Luigi Gubitosi; and our CFO, Giovanni Ronca. As usual, Luigi will provide an overview of last quarter's achievements, and Giovanni will illustrate our financial results. Pietro Labriola connected from Brazil will participate to the Q&A session. And pointing out to you our safe harbor disclaimer on Page 1. Let me hand it over to Luigi. Luigi, the floor is yours.
Thank you, Carola, and good afternoon, everyone. Good morning for those of you connected from the U.S. The presentation title summarized one of our key missions. We have taken the leadership of our country digitalization, and our contribution increases everyday, be it through the broadband or through digital services for consumer enterprises or public administration. We see it very clearly in our quarterly results, action and initiatives. We keep improving the quality of our services, which is reflected in growing CSI and RGS profile, which is rewarded through increasing number of ESG indexes, such as the last one launched by the Italian Stock Exchange last week. TIM workforce is becoming leaner, and at the same time, we're recruiting in our digital companies. In Fixed, we launched a football offer as exclusive distributor of DAZN. This is supporting churn reduction and the acceleration of ultra broadband access growth. Indeed, in the first 9 months of 2021, we almost doubled our fiber net adds compared to the first 9 months of 2019. We are pleased with our mobile performance in the quarter, another record low churn and service revenue improving strongly quarter-on-quarter. And cloud revenues are growing faster than expected. In Brazil, we recorded another quarter of growth, and we expect a number of positive catalysts for Q4, starting from the expected approval of the Oi transaction and the closing on the FiberCo deal. As you see in Slide 4, group service revenue trend is improving quarter after quarter and is now almost flat year-over-year. Net debt is down more than EUR 3 billion year-on-year. You may remember that in the last quarter, we focused on 4 drivers that are contributing or will contribute to our growth. Some of them are startups within our group. I recently said in an interview, we are transforming TIM in a proper TMT company, some are sector on macro trends. Number one amongst the tops, football as a complement to our convention strategy. Customers are switching from satellite to fiber. We expect this migration to accelerate, considering the football viewers are almost 5 million with over 3 million Pay TV that were indeed mainly on satellite. The rest are pirating, and they will have harder and harder times going forward. I promise you. Second growth driver. We continue to see the Italian fixed market growing, with mobile-only customers acquiring fixed as well or going back to fixed. Third, growth driver beyond connectivity. All our digital companies are growing double digit, adding towards their target of more than doubling by 2023. They accumulated EUR 700 million revenues generated in 2020. Fourth, the recovery and resiliency plan and the related improving macro context. Let's see where we stand on each of them. As you know, football was launched on July 1, making TIM Vision, the richest TV platform in Italy by far. We are currently offering Ceria, Champions League, Disney, Netflix in the same package with underlying price of EUR 44.99 after an initial promo period. Although revenue impact is negligible in Q3 due to the summer promo, the benefits on our KPIs are evident. Much lower churn reverses typical negative Q3 seasonality. It is the first time ever that fixed churn falls in the summer quarter, further boost to our ultrabroadband net adds already pushed last year by the pandemic and, of course, an increased growth in TV customer base. And importantly, this growth opportunity led by football is not just for us, but for the entire country to boost digitalization. As I said, and I'm now on Slide 7, football is a complement of our fixed to fixed strategy on which we continue to push hard. Our ultrabroadband coverage was further expanded, now reaching approximately 94% of active lines, and in fact, proved to be very solid even when tested against football. We're growing on convergence, thanks to TIM Unica and digital payments, both supportive again on churn reduction. We're growing double digit in ICT and reached 25% year-on-year growth on cloud. All these efforts together are contributing to the country fixed market growth, which in Q2 for broadband was the fastest among European operators. Again, a first time. Third growth driver are digital companies. We want to shed more and more light on them. Noovle is running above targets with a 25% growth in Q3, adding towards reaching and most likely topping the target of EUR 1 billion revenue and EUR 400 million EBITDA by 2024. Recently, Noovle signed important partnership with Cisco to improve our cloud offering for the public administration; and with Oracle, for the development and joint offer a multi-cloud solution to Italian enterprises and public administrations. Sparkle is growing double digit, and it is investing in profitable infrastructure that we have huge demand for as well as in growing opportunities in the Enterprise segment. I expect Sparkle to approximate EUR 1 billion revenues already in 2021. Olivetti and Telsy are growing double digit to and investing in additional future growth and value enhancement. As an example, Olivetti has acquired last summer, 9% of the listed company cycle. And in a few months, SECO doubled its value. Thanks to this industrial and financial partnership, Olivetti and SECO, will jointly pursue opportunities in the IoT markets. All these growth rates are obviously a good explanation of why the market pays for these assets, multiples well above telcos. On Slide 9, we are zooming on our digital companies business, and you can see Noovle sign over 700 contracts in Q3 or for an annual contract value of around EUR 75 million and reaching year-to-date 1,800 signed contracts. That's quite impressive of what we can still call a startup. Clients acquired range from Credit Agricole to Generali and from the military arm, Carabinieri to SMEs. And indeed, we are building a wide and compelling offering from pure infrastructure to cloud managed services, to professional service, analytics, business application and security for our clients' needs. We are digitalizing Italy cities or places, be it offices or even landfills and also Italy's jewel, it's touring industry. It is an effort that is already paying off. Without Noovle, we wouldn't have had the opportunity to participate to the largest cloud opportunity ever in our country, the National Strategic Hub. The Italian government has made a call for proposal with the objective of hosting and managing strategic public administration data and services. We have responded together with CDP, Leonardo and Sogei. These 4 companies will create a Newco, where TIM will have a 45% stake. The Newco will be offering cloud services and infrastructure to the local and central public administration, in turn, buying services, mainly from industrial partners. If our consortium is awarded a contract, which I feel confident on, TIM will be providing the cloud infrastructure and the cloud services; Leonardo will provide security services; and Sogei, training. CDP Equity would participate as financial and institutional partners and has been the organizer of the consortium. It is a huge initiative, very important for the country development that will be fueled by the recovery and resilient plans for about EUR 1.9 billion over a 10-year period for the setup and migration of the public administration to cloud. On top of that, the Newco will learn from recurring revenues related to the services provided to the public administration, which is likely to be over time an even higher source of revenues for the Newco and for its Partners. We're now on Slide 11. I mentioned, in fact, the recovery plan. We are showing this slide a detailed list of projects with a clear telco or digital component and my applied TIM. As you can see, telco-related projects are not only the ones in the digitalization missions. The short list alone as the value of resources allocated in excess of EUR 43 billion, and we have task forces working full time to build our contribution and participation to this extraordinary journey, which will bring Italy to be fully digitalized. Before leaving the floor to Giovanni, let me recap where we stand relative to Italy's telco market. This indeed is an important slide. Fixed market is growing, and we believe it will keep growing as there is still space to in back the mobile-only customer base and families that do not yet have a fixed broadband line. In this context, TIM is increasing its leadership. It is a crowded market, we have many competitors and some have done the easiest thing. They moved down prices. This is quite irrational considering that we sell an inelastic service. This dynamic has been going on for some time, and we are now the cheapest telco market in Europe. We believe it's time to change and that the way forward, service excellence and quality. We believe people realized how important their speed connectivity is and they are ready to pay a few euros more to have an excellent service. We're not simply raising prices. We're raising our ambitions to make this market healthy for the players and better for the customers. And so there you go, Magnifica, our new tiered offering. We are redesigning our fixed offering with a clear strong push on the top of the range tariff, Magnifica, which at the price of EUR 49.9 provides a speed 10x higher up to 10 gigabits per second, the best of our WiFi. In fact, a top of the range WiFi 6 and security services as well as dedicated caring and priority access to our stores. So in short, you can add the top of technology and the top of services, and we'd refund those that are not happy with our service. But they will have an excellent service also with our executive and premium offer. Together with our convergent offer, we continue to segment the market in order to serve differently, the different clusters, with different requirements and willingness to pay for service. Now let me hand it to Giovanni, who will illustrate our quarterly financial results. Thank you. Giovanni, please?
Thank you, Luigi. Good afternoon, everyone. Service revenues are on a progressively improving trend and decreasing minus 1.4% year-on-year in Q3. Group EBITDA after lease was in line with Q2, with domestic affected again by a few discontinuities. First one is higher advertising and commercial costs related to football and start-up costs for the group's digital companies. The second one, Q3 '20 COVID related savings are coming back with a tougher year-on-year comparison on cost this year, mainly on indirect labor and real estate. Thirdly, impact on EBITDA of lower equipment and handset sales with slightly lower margin mix versus last year. We can say that like-for-like, net of these discontinuities, EBITDA after lease was down 2.4% year-on-year in Q3, similarly to the minus 2.5% year-on-year in previous quarter, helped by the strong fall in labor cost, down 13% year-on-year. Let's move on Fixed in the next slide. Fixed KPIs continue to be strong. Retail lines are again broadly stable quarter-on-quarter with a much lower seasonal slowdown versus last year. As Luigi pointed out, retail ultrabroadband net adds totaled 652,000 net adds in 9 months, the highest level of the last 3 years. Churn was reduced again quarter-on-quarter despite seasonality to a new record low, first time ever that churn does not grow in Q3 versus Q2. Including wholesale lines, ultrabroadband customer base reached over 9.7 million lines with an 18% increase year-on-year. Moving to the next slide. Fixed service revenues were down minus 2.5% year-on-year, improving year-on-year performance. 1.2 percentage points versus Q2 net of 1.5 percentage point quarter-on-quarter swing in ICT contribution to growth due to seasonality and delay in a couple of contracts. The trend is explained by international sales growing double digit, showing the good work that is being carried out in Sparkle repositioning, by national sales improving the year-on-year trend versus Q2. And finally, also retail helped by customer base stabilization substantially offsetting the ICT swing described above. Customer base evolution still represent a 1.8 percentage points drag year-on-year, but improved 1.1 percentage points quarter-on-quarter. Total fixed revenues were impacted by lower equipment sales with Q3 2020, benefiting from post-pandemic rebound. Next slide. Mobile KPIs improved alongside the continuous development of our 5G network, which has been awarded as the fastest in Europe by Open Signal. Importantly, the mobile number portability market was down 31% year-on-year to a level of 2.3 million lines, which is the lowest in the last 10 years, meaning that our rational approach to pricing has helped to cool down the washing machine effect. For TIM, net adds almost doubled quarter-on-quarter, benefiting from the continued good performance of machine-to-machine and IoT business with human net adds losses almost half quarter-on-quarter. Churn at a new record low level over the last 14 years. I'm on Slide 19. Mobile Service Revenues trend improved 4 percentage points quarter-on-quarter as one-off tracks faded away as we have anticipated in Q2 results call. ARPU Human improved progressively and actually increased 0.9% year-on-year, net of the drag from CSP cleaning that, in any case, is also fading away. Total mobile revenues were impacted by lower handset sales due to rebound in Q3 2020 after the Q2 full lockdown. Slide 20. Addressable costs were down year-on-year, helped by lower commissioning and labor costs with the latter benefiting from a full quarter of solidarity, lower holidays and our continuous FTE reduction. On the other side, we had few drags already mentioned, namely football and digital company startup costs and the COVID-related cost saving rebound. Variable costs were up year-on-year, mainly for higher interconnection related to Sparkle sales growth. Moving to Slide 21. Let's talk about CapEx. Luigi said, we are investing for growth, and we are doing it on a timely basis with CapEx and OpEx differently splitted during the year versus the past. We anticipated in the first 9 months, the CapEx that was normally concentrated in Q4 with consequence both on the size, much higher in the first 9 months and lower in Q4, and on payments, more skewed in H2 rather than in the following year as was normally the case in the past. Importantly, we are doing what we call the good CapEx, i.e., CapEx for growth, such as the acceleration of the FTTH rollout as well as our digital company's investment and football. The consequence of what I described is that Q3 working capital and equity free cash flow was impacted by higher CapEx and higher payments of H1 CapEx. Moving to Slide 22. TIM Brasil reported solid results amid softer macro recovery and despite the highly competitive environment. Service revenues grew 4.2% year-on-year, supported by mobile postpaid and continuous growth on Fixed driven by TIM Live. The strategy to focus on customers' value continues to pay off and is leading to increasing ARPU, both in mobile and fixed. Mobile ARPU grew for the 23rd consecutive quarter. EBITDA grew 4.4% year-on-year with 0.7 percentage point margin improvement. TIM Brasil continued to create new sources of revenues through valuable partnerships and confirm its commitment on ESG in line with the rest of the group. With that, I leave you to Luigi for an update on strategic initiatives and his final remarks. Thank you.
Thank you, Giovanni. Starting from the strategic initiative, we want to move forward in the industrial transformation of the group with the objective of designing and implementing, also to carve-out projects. TIM optimal configuration to face the challenges and pursue the opportunities offered by the new digital priority. You will see in the near future a leaner more decentralized DI with focus on each and every one of its businesses, of which we intend to fully exploit the value. As you know, we announced INWIT value to the merchant synergies with Vodafone tariffs, and then we partially monetized the asset. We'll do the same with Noovle where we will follow a double track of a listing or letting investors in the capital of the company with a minority stake, and we see many other opportunities to follow a similar path to additional carve-outs. Let's move to Slide 26, guidance. We'll give you a couple of important messages today. We reiterated our strong way to transform TIM in a proper TMT, which is leading towards growth, but some step-up costs to absorb in the meantime. Secondly, in order to face the market condition described earlier, we decided to selectively raise prices in Fixed after experiencing that a rational behavior in mobile did not, in fact, impact our KPIs. We believe the market is ready for such a move and the rationality will prevail. We are confident these are the right choices that will deliver the growth rates we have projected for the coming years. Short term, we're paying what is needed to reach such growth, both on CapEx and on OpEx. We need to find ourselves ready to capture the opportunities that the Recovery Fund and the digital world offer us. The Cloud National Hub is a good example. In order to cater for all these potential setup cost, and a different phasing of our CapEx during the course of the year compared to the past that as Giovanni described had some impact on working capital, we are lowering our guidance range for Group, Domestic revenues and EBITDA and the cumulated 3-year cash flow. The second reason is, of course, related to the market conditions, as we described before. We are totally committed to do the change. As you saw, we started to act and we'll continue to bring as much rationality as possible in the market. While we've updated assumption on the Brazilian reais to reflect recent market moves, we remind that expected upsides from the Recovery Fund is still not embodied in our guidance, neither is the acquisition of Oi mobile assets. So a few final remarks on Slide 27. Group service revenues are almost stable. Domestic fixed lines were stable for the last 4 quarters with ultrabroadband growing fast Mobile and fixed churn are at record low levels. So we performed better than most in a difficult market. However, deteriorating pricing levels did impact our expectation during the year. So we are now selectively increasing prices, and we believe the rest of the market will do the same. Our portfolio digital company puts us in a unique situation to benefit from the National Resilience and Recovery Plan on cloud network, 5G and digital. So despite some delays, we remain optimistic on aspect of it, both on the macro and on TIM. Lastly, TIM needs to focus with more emphasis on reducing its cost base, decentralizing structure and pursuing all the market opportunities. We're investing in new business behind connectivity to achieve growth and create optionalities. And we are very committed to extracting value from our unique asset portfolio. With that, let me open the Q&A session.
Operator, we are ready for the Q&A. Thank you.
[Operator Instructions] First question comes from Mr. David Wright from Bank of America.
I've got so many, but I'm going to try and limit myself to just a couple. I just want to understand the guidance a little more, please, Luigi. The fact that after guiding down in Q2, you're guiding down again in Q3, is this purely a function of the market pricing right now? And is that the decision behind your raising prices? And I guess my question or, dare I say, concern is, do you need that to work within the '22, '23 guidance? Do you need pricing to move upwards? If everyone else doesn't play ball, could we see some incremental risk? And then I guess just on the Fixed revenues, having been so strong for so long, the ICT was much slower, and you mentioned some project delays. Is that just phasing? Should we expect those projects to come back in Q4 and growth to be back in the 20s? And is that the outlook for 2022? Do you think is it just phasing there that meant that, that fixed line revenue was a bit weaker? That would be very useful. And I don't know, can you give us any indication on net additions in football as well?
Thank you, David, and thank you for asking the first question, which allow me to answer, and I have to say, I'm very disappointed with the fact that we did have to change our guidance yesterday. The reason is the following. Going back in July -- there is some disturbance. Okay, now it's okay. Going back to July, basically, what we did was mathematically to take into account the fact that there were no voucher, and there was no -- the soccer, there was not -- has not been included the cost of the -- the football, which instead we had to do. So we did not do a full forecast with basically some bullish expectation based on July performance. And in retrospect, that should have been challenged more, which we did not and I regret, and that was perceived as a change of guidance, and it was to a certain extent. But I think we did not explain enough. There was a sort of mechanical. And our bullish expectation that followed a very, very strong month of July, basically, we did not perceive well enough that the market was deteriorating in terms of pricing and that we would have had some additional expenses for, I mean, -- the good part is that we've been investing on the "other businesses". And we did not try to, how could I say, optimize a single quarter, but we're managing with a long-term view. So the answer is we should have done all at once, and I agree and regret that, that did not happen. It was a mistake. We should have challenged those expectations that we had at the time more aggressively, internally. As I said, we regret that it happened. But -- so you shouldn't take it like we did exercise twice. Unfortunately, we did exercise once. We did it now with the results as of September 30. We gave time to adjust also to new people that was managing the commercial side. And as I said, we were expecting a trajectory that was more strong. And at the same time, we were not expecting, frankly, the fact that some of our competitors lower price very, very aggressively. Frankly, this is just a -- the unusual things about this sector these days is that we are selling an inelastic service. So doing lower prices basically means that the entire market, at the end, is worse off. And that brings me to my last question. Yes, indeed, price is very important, and we saw that. But price is not the only service. It's not the only element. And in fact, there is a part of the market that it's extremely price sensitive. And I think in mobile, basically, now that the cluster is very well identified. In Fixed, it's -- it is even more important to segment and this is basically our new policy. So we have raised prices -- more than raise prices, I think, what we're trying to do is to segment the market and to cater different clusters. There has been a significant study about the market. We're actually making more use than usual of the AI and data analytics to do these moves. And we believe that will be successful, which does not mean that everybody will migrate to the upper segment. I mean if you look at in other industry, you don't have -- if you take a British Airways flight from London to [indiscernible], you will find that the number of seats in business class are not the largest number, but there is a number of the customers that do take it. I envisage that our ability to price will have to become more and more like in other industry, less mass and more targeted. I mean we need to profile more customers. Basically, we are pioneering as a helper of customers in the cloud industry, in data centers and artificial intelligence. And we're now trying to apply the same principle. So we're quite convinced. Now your question is what happens if other competitors do not raise prices? Well, I know that can't say the long run, but in the long run, they are all dead. What I mean is that -- maybe we are all. What I mean is that you're now seeing our results, and some of our competitors do not have the privilege of being listed. But I'm not sure that anyone is doing better than we are on a relative basis. In fact, if you look at our KPIs, they're extremely strong. Actually, while we were coming to this conference, I think, AGCOM came up with their second quarter numbers, and we've gained again market share. It gives TIM Retail plus 4 points compared to last year in FTTH, plus 5 points in FWA, and 0.4 in FTTC, which, as you know, our market say is already very strong. So in a P times Q equation that gives us the revenues. The Q side is becoming very well addressed. And in fact -- and I agree that you have to do both sides of the equation. You also need the P. But if I look at the churn, I look at the CSI, I mean I look at the things that are happening, I'm quite happy with our performance on the Q side. Clearly, the price and the deterioration in the market prices does not depend only on us. But -- and we do not talk with our competitors, of course, about this. But my perception is that they should follow. That's what a rational player will do. If they want, we'll -- we always have Kena. We have our premium -- don't forget, our premium offer, so the entry level, so to speak, of our 3-tier offerings. It's at EUR 29.99, which is basically where our standard offer is now. What we're basically offering, it's for the first time in this market, a personalized offering a service. I mean, you will have a dedicated group. You will jump queues, you basically -- for a relatively low price because we're now focusing on every single cent in this industry. But if you look at electrical tariff, for example, there are many folds higher. And basically, people don't complain. Or other utilities is the same. We have the cheapest utility, I guess, on the -- on earth at the moment, telephone companies everywhere, in particular, Italy, are cheap. And as I said, some of the things that you're seeing, for example, the number of -- the fact that churn decline basically means stability, means less commissioning and so on and so forth as well as the fact that people are moving more and more to the fiber, will mean that for us, there's going to be less copper in the future. And I think we are starting to discuss end of sale in certain areas and phasing out copper more aggressively, which in turn will reduce cost. I think I was a bit long in elaborating my answer. But the question is, yes, we see prices higher. But of course, we are market players, and we react to market movements if -- as we see them. And as I said, going back to the July exercise, that was purely arithmetic, and I think we explained that. We tried to explain at the time. I don't know if we succeeded. Then we had to take into account other facts that, at the time, were not as evident or maybe we missed the signals at the time. I hope I answered your question. And you said you have many. I'll be happy to take others or maybe if Giovanni...
David, I think you asked us a second question related to the trend of fixed service revenues in the rest of the year, in 2022. Let me stress a couple of points. On Q4, we expect the fixed service revenues to be substantially stable. Please remind that in Q3, the customer base evolution was still almost 1.8 percentage point drag on revenues. And this will progressively fade away in 2022. So looking at 2022, I think there are a few factors that will support the growth of fixed service revenues, for sure, football. ICT services, in particular, cloud will be a very good service that we are going to sell. Additionally, Sparkle will continue to perform. And on top of that, the customer base stabilization will support the overall picture.
By the way, David, if you have other question, we'd be happy to follow later on with Carola, as usual.
Next question comes from Mr. Fabio Pavan from Mediobanca.
I would have many, would try to focus on 2 topics. The first one is the strategic initiatives or the actions you are willing to implement to crystallize the value of your assets. Could you provide us some more color on this? There is already a turning table, cloud could be the starting point. And in order to focus on the cloud, on the national cloud hub when we should have some visibility about eventually the awarding of these contracts? And is it fair to argue that this Newco will actually become a client, a customer of your Noovle factory on the cloud?
Indeed, in fact, let me start from Noovle. You say, when we could make it actionable. And actually, it depends on PSN in a sense that. The PSN, it's huge opportunities for everybody. It's a huge opportunity first of all, for the country because it would be an opportunity for the public administration to accelerate dramatically their process of modernization. It is also a great opportunity for those that work in the field. And we feel that we have a particularly strong group of players that team up together. As you know, there is a consortium headed by CDP, which comprises ourselves, Sogei and Leonardo. Now we -- obviously, I cannot comment on when it would be awarded because there is a process. I don't -- I cannot represent that would be awarded. I'm very confident -- But this is something that you would have to ask the government. I would imagine that the government is saying that they want to put the money at work for the Recovery Fund has up and they have 5 years from the beginning of '22. So I would imagine that at some point at the beginning of '22 that, that should occur. Now why did I say that it's relevant because this new company will basically cater to the local and central public administration entities. And we'll use suppliers of some of the, let's say, the most relevant component of its purchases, the -- its shareholding base. Obviously, not CDP, which is the sort of promoter of this consortium, but it's not a technical or industrial pattern. It will use ourself, Leonardo and Sogei each for different type of services. So I would like to incorporate that in the business plan that we could present to potential investors in the company. So not only as an option that clearly everybody discounts, but as a fact, which does not mean that we should not start preparing in that respect. And anyway, and actually, I'm happy to report, and you saw in the presentation that we're going at a faster pace than the one that we originally told you when we start talking about Noovle. When Noovle was first mentioned at the time, I don't know if even the name Noovle existed. But Anyway, we were expecting to double our revenues in the field, which in 4 years, which, as you know, imply a 16% growth rate. Now since then, has always been -- the first digital who has been there, too. And in this case, I mean, in this quarter, I think we got to 25%. What's even more encouraging that besides the partnership with Google, which remain in the private side of the business, not with the PSN. I'm referring now to the Noovle characters, the typical business, remain a milestone. We have made some other important agreement. We have a partnership with Cisco. We have an important partnership also with Oracle. And we're working with others. And as I said, we were becoming stronger in the cloud by the day, I would say. And that today the multiples of the cloud make it very encouraging. We have made significant investment, but in [indiscernible] we have now 1,800 contracts signed. And we were with prestigious -- in Q3, we got people like Credit Agricole and the Carabinieri themselves, which is -- make us proud that we can help our -- the people that help us, help us Italians, in general. Generali, Azimut, so -- and many others, and we will want to extend more and more to small and medium enterprise. And this is before any effects from the PNRR. We do expect when a PNRR starts and allows subsidies for migration to the cloud, this indeed is going to be a very interesting business to be in. So to cut the long story short, because you're going to ask me a date. I would say that you'll be probably going in the second half of '22. Again, we probably start at the beginning of '22, the proprietary work anyway and expect that in the first half, the PSN issue get resolved by the government and awarded to somebody, which we hope and expect it to be ourselves. I mean, our consortium, obviously. And then we'll see how we can maximize the value of the company. More in general, I think that, first of all, let me reiterate that we obviously are an industrial company and look at how we can strengthen our industrial base. And so the logic is going to be INWIT like or FiberCop like. And by the way, in retrospect, we're very happy with both moves that we made at the time. Clearly, most of our value, at least today, I think, cloud, which has become part of -- very much part of our core business will surprise us positively in the future in terms of valorization, but we have a lot of value in the network -- or I should say, in the networks. Now the FiberCop is separated from the primary. And we will want to capture part of that. I mean, we want to make sure that the market does realize that and see if there are opportunities in that respect. And before you ask, I think, I said and I reiterate that we do share a good relationship with CDP historically. And with new management, we both have, I think, I say then reiterate a pragmatic approach with a desire to do and achieve things. And so as soon as Enel leads and CDP becomes in charge, and so all potential bottlenecks are removed from the discussions. We will reignite a flame that never got distinguished. And basically, this moment is arriving because from my understanding, November is the month in which Enel should be out. And so time will tell -- but as I said, we have very important assets in the field, which have become stronger since we spoke, as a matter of fact, since we spoke the first time about this type of issues. Over the last 3 years, technological team has increased strength. By the way, once 1 giga was mentioned by somebody as a proof -- future-proof technology. We're now marketing 10 giga for everybody as you've seen in our offering. So technology moves very quickly. We're happy to be at the forefront of these moves. And I could speak further about assets and so on and so forth, but I think it will be appropriate that on certain issues, I speak with the Board first. So what I got yesterday was an encouragement to continue exploration around this concept and then to report back about the opportunities that we have. I hope I've answered your question.
Next question comes from Mr. Mathieu Robilliard from Barclays.
I had a first question about your guidance for equity free cash flow. So you lowered the EBITDA guidance for 2021, but that's unchanged for '22 and '23. And also you left CapEx unchanged. So I guess a rough estimate would be that maybe you lose EUR 200 million for the equity free cash flow because of the lower CapEx, but you're cutting the guidance by EUR 0.5 billion. So I was wondering what explained maybe the difference and also what a change there may be from Q2 in terms of the free cash flow generation. So that was my first question, please.
Mathieu, Giovanni speaking. I think that the evidence on equity free cash flow is already in what you can see on Slide 21 in the quarter. So the fact that we did some good incremental EUR 160 million CapEx and that the payment cycle of this CapEx was different from the past. I mean we anticipated them in the first half of -- a few CapEx in the first half of the year with the cycle payment -- with the payment cycle of the CapEx, you get the payment in the third quarter and not later in the year. So this kind of anticipation, higher CapEx compared to the past and a different payment cycle is creating an effect that is justifying the shift in the equity free cash flow also going forward. Then you have an FX component that is important. As you said, we updated the real exchange rates. And you have the EBITDA effect overall. So this explains the mathematic for the revision of the medium-term CapEx equity free cash flow projection.
Let me add a comment, if I may. I'd like to add a couple of comments to what Giovanni said. First of all, this year, and we told you whether a startup that requires some investment. Now this investment added a different timing and a different nature from the usual ones in the sense that we're happy with our guidance, our typical guidance, and that's enough for us to develop what we need to do. This year, where this opportunity of football on which we remain extremely committed, in a sense that 3 years from now, hopefully, even 6 months from now, this will have proven a very good move for our company. And we took the opportunity and, in fact, we needed to, was both a need and a willingness to strengthen our network. As you know, there were concerns, also expressed by many before we started with football, about could a network manage such a flow of data? Could view rate go everywhere? Are there going to be people that are not going to be able to watch them? Well, this did not occur, which does not mean that there were not some technical issue. And start here in the place to explain exactly what happened and where. But I can't tell you that whose responsibility that was, but I can tell you that there were not network issues. In fact, the network performed very well, have been performed very well after the pandemic. And I dare say, well, that we have the largest network in Italy. It, historically, is a fact. But I would say that we have one of the most advanced in Europe. We are the only player in Italy that uses multicast, which is the best way to watch a football match, by the way. We did develop a certain specific adaptation of the technology. And in fact, that's appreciated by OTT other than DAZN is becoming steady. So my point is that we did -- we always do manage with a view to the long term. And this year, we increased it because of this. But we go back to -- I mean this has been done. It works. We are fine with that. We go back to the previous guidance from next year. The other thing is that you will see you will see that we remain committed to generate cash flow and debt on both sides So -- and I know that it's -- every time that there is some revision or the market changes, one tends to always re-question everything, I mean, the analyst side. But not on this side. We will continue to generate equity free cash flow, and we continue to reduce debt, and we have not revised our objective of bringing debt down through both organic and inorganic. We remain firmly committed. We would like that in 2023, this occurs and, in fact, as it would coincide with the end of the term would like to leave it as a lasting legacy. I hope we have given both the answer and some color around it, and we will move to the next one then.
Next question comes from Mr. James Ratzer from New Street Research.
Yes. I have 2 questions, please, Luigi. The first one is just coming back to DAZN. Could you please give us an update on actually just how many customers you do have taking the service at the moment? And in particular, if you can give us any steer about how many of those customers are brand new to the Telecom Italia network. And it's also been, I think, written in the press that you'd have a minimum payment to DAZN of around EUR 336 million per annum. I mean, how many customers do you need to get to be able to cover that minimum payment? And then the second I had was, just would love it if you could please give us an update on the voucher scheme. It seems like half of some initial successes that project has, stalled during the past 2 quarters. And I think only about half of the Phase 1 vouchers have now been awarded. Could you please give us an update on how you see that progressing? Should we expect a pickup from that in Q4 and later thoughts on the timing for the Phase II voucher scheme?
Okay. First of all, on the football. Now with regards to specific numbers, our agreement with DAZN prevents us from disclosing any figures, but let me try to explain it on a qualitative basis. The press, as you know, writes lots of things. Some are correct, some are not. But not necessarily accurate or fully accurate what you've -- you're right in a sense that, yes, there is a minimum guarantee, but there's a number of other clauses and let me do the bottom line. I don't expect at this stage to have an impact this year. So as we speak today, we are going, as I said, slower than originally thought, in a sense that we had expected quicker takeup. We had a very strong July, and it now has started picking up again, but August -- second half of August and September were somewhat less effective than we originally thought. Now we are marketing the football together with Netflix, Infinity, and we have a bundle we call Tutto, all basically, which has restarted some more interest in the public. And I am confident that -- we are confident, I mean, as a team that we're going where we should be, and we will reach our objective, although a bit later than originally expected, also because we are fighting against the one stronger for us in business, which is inertia. I mean, if you had certain habits for 20 years, it takes longer and maybe there are also actions that have slowed down this process, including the need to have a complete awareness, some information that -- there have been a number of factors, but the trend is clear. And the trend is not reversing. I mentioned to you that we gained market share and the churn has gone down. Now I don't know yet the same figures for our competitors, but I do expect that this has played a significant role. The more customers we're getting as Unica or bundle of quadruple play, triple play and so, the more they tend to be more resilient, and they stay with us more and -- looking simply at the chart and you see that in Q1 2020, it was 4.7 the churn, and today, it's 3. That for me is a matter of major satisfaction. I think you were also asking about the vouchers. I'm reluctant to speak for the government because they always say something -- my understanding that the Phase 1 of the voucher, it's basically almost ready and almost is always what makes the trick. I think they're really finalizing it. We will ask if they can clarify that -- I mean, also as an association, I think we and the other operators are quite interested in knowing when it's going to be the start date. My understanding is that it should have been cleared by the European Union or just about. And therefore, they just need to finalize the launch. So when it's happening? The exact date, I cannot tell you. We'll try to see if we can ask the ministry to give us some official. And the amount, as you know, is EUR 900 million. And my understanding is that there's going to be a first phase, which will launched immediately or just about, which is going to be on the small and medium-size businesses and it's EUR 550 million. The residual will be for consumer and to start at the beginning of the year. So this is basically our state-of-the-art knowledge or better. It's an educated guess.
Can I just follow-up just on the churn point. I mean looking at the consumer broadband net adds were, I think, 38,000 this quarter. They were 22,000 this time last year. And you're talking about the kind of churn reduction coming down as well, which I think could be worth about kind of 5,000 extra support. I mean do you think then the gross adds pool in the market hasn't grown by as much as you were expecting. I was just trying to figure out why we haven't seen a kind of stronger pickup in overall broadband net adds.
Well, I'm not sure I've understood your question. So let me try to answer. You're saying whether, in my opinion, there should have been a bigger migration to ultrabroadband. Is that your question?
Well, the more about just -- I mean maybe I'll take it off more about just overall size of gross adds available in the broadband market in Italy. I was expecting to see that growing more with the voucher program, fixed mobile migration, DAZN. But it seems as if the gross add pool available in the market might not have grown much compared to last year.
Well, okay, first of all, the voucher, as I said, DAZN really started the large voucher program that is supposed to happen. The broadband lines actually went up by 600,000 in 1 year. So if I take -- and these are the AGCOM numbers, I suppose, the time being passed, there were EUR 17.8 million in Q2 '20 and now EUR 18.4 million in Q2 '21. And arguably could have been higher? Yes. And they will be when the voucher program will start. But I have another -- this is on Page 7 of this presentation. You see that, actually, Italy is leading in terms of broadband net adds. So I don't see it as negatively as you are. In fact, if we can have the slide on -- okay, it is on -- I think you can visualize the market has increased and the broadband market has increased more than the fixed. So again, there are migrations. And as I said before, the earlier we do the migration, the better for us and the market. But Italy is actually leading in Europe. I guess for now you will see that with the voucher scheme that will accelerate further. So the pool varies from time to time. Also you should take into account that there is a strong correlation between mobile prices and fixed that we also run. I would say that if we see the market reacting to the vouchers well, the pool will significantly increase. And we expect that our strong performance of market share and net adds will continue.
Next question comes from Mr. Giorgio Tavolini from Intermonte.
A couple of questions on my side. The first one on cost initiatives. I was wondering if you expected to take additional cost cutting measures to support EBITDA trends in the coming quarters? And the press recently talked about some additional actions on labor costs, in particular, if you also see any inflationary pressure on energy cost, especially looking at cloud, that is a quite energy-intensive business. The second question is on CapEx cycle, a follow-up. You had a significant working capital absorption in Q3 for the advanced payment of CapEx that was initially expected in Q4. Should we expect a complete relief in Q4 or maybe an additional absorption of advanced payments of CapEx expected for 2022?
Okay. So Giorgio, I'll start with OpEx and then Luigi will elaborate on energy and I'll address also the topic on CapEx. So OpEx for Q4, I mean, our conservative estimate is that they will be in line with year-on-year. You have there 2 components: The first one is the path in terms of labor cost reduction will continue, so it will continue to go down. On the other side, you will have the continuation of investments in football and cloud and other products that will -- is part of the plan. Obviously, everything we do aims at rightsizing the construction of the company. So this continue to be of paramount importance. You were asking about the payment cycle of CapEx. The payment cycle of CapEx is twofold. You have higher CapEx compared versus previous year. And then CapEx that have been realized in H1 and paid in Q3 and Q4. So this is what you see on Slide 21. We do expect to have a fourth quarter in which from a working capital perspective, you will have the cash generation. So what you are saying is true. Obviously, considering the overall path of the year, so we do expect that Q4 that will be positive in that respect.
Yes. And I would like to add 2 comments to answer your other questions. One on energy. I think the impact on '21 is limited because we tend to hedge in advance our usage. And I think we're basically hedged for the year. There is a very limited amount. Giovanni?
We have a 97% hedging as we speak of the 2021 consumption.
So I would say '21, the variance can be very limited either side. I think we have hedged most of next year as well. So obviously, there is still a component. We tend to think that energy prices tend to be at a relatively high level? Well, they are, as a matter of fact. We think they're near peak somehow, but as in general, we tend to hedge as a policy. Bear in mind that in cloud, depending on contracts, but I think in most contracts, energy is a pass-through. So you tend to pass the cost to the customers at the end of the exercise, charging a small fee on it. But if data center itself absorbed, but then some of the costs are borne by the clients. And then there was another question. On the cost base more, in general, is that -- well, I guess you always look at the cost base. And if the price side tends to be less favorable that you had originally anticipated or you need to look at the cost with further higher degree of scrutiny. And without entering on specifics, we expect that in the next plan, there's going to be a more stringent approach to cost and that will reduce further the base. I think we did not answer one question. I think Mathieu of Barclays asked, if I understood correctly, Mathieu, you wanted to make an additional question, and we did not let you do that. If that's the case, can you please get back online and ask your question. We apologize for that. Okay. Then let's go with the next questions.
Next question comes from Mr. Andrea Devita from Banca Akros.
I was looking at your statement on the EBITDA that there is a 3 percentage point or EUR 40 million impact from football and start-up costs. So I was just wondering whether these start-up costs are something that is going to be repeated in the next few quarters and maybe in 2022, as I was thinking that with the revenue growth like in Noovle and Sparkle and Olivetti, there was some EBITDA improvement and not decline. So I wonder whether these start-up costs are all related to football or is, in general, a more investment needed in the other group of the subsidiaries. This is the first question. The second is just if you have stopped giving an update not on DAZN subs, but just on TIM Vision subs. So I just see a qualitative histogram, sorry. I wonder whether there is a specific number on that?
I'll start, Luigi, on the EBITDA side. I mean, there is a breakdown on Slide 15 on EBITDA with the 2 components. I mean, the decline is related partially to, I mean, for more than half to the rebound of COVID-related cost savings and the sale of equipment, the lower sale of equipment and handsets. So the rise to 3 percentage points are related to the investment in football and the start-up cost of digital products or digital factories. And I think that the first portion will be certainly a one-off. The rest is part of the development of the plan.
Yes. Let me give you some more color on that. As a matter of fact, the company that has absorbed more CapEx has been Noovle, where we have basically 3 data centers, to hyperscale in preparation in the Turin region, we have 3 in Milan, and 1 in Rome. Basically, in Rome, we had already some other infrastructure that we will use alongside the one that was talking about. Basically, most of them, if not all of them, will be in service. So work will be completed by the end of Q1 or beginning of Q2 of next year. So you will not have the same type of absorption in terms of cost to add this year. As a matter of fact, we expect them to start generating revenues in a very significant time. At the moment, they will start -- they will be obviously amongst the most -- the newest and most seasoned in Italy. So with the other companies, the level is low. In Telsy, we are investing because in the business they do, this is -- they're quite generative, and so it's a profitable business. But they -- we would like them to enlarge their spectrum. The same is true for Olivetti. And in the case of Sparkle, I think that their main project is so-called Blue & Raman, this cable linking Milan through Genoa and Palermo and then the Middle East to India, which is proving to be extremely, extremely attractive to new customers and business partners. That -- a significant chunk of this investment has been done this year. There will be some also next year. But I will say that all in all, with these companies, most of it has been -- most of the CapEx has been related with Noovle and has been done this year. I hope I've given you some infos about -- I'm sorry?
No, just to be sure whether we were talking about EBITDA and whether it is growing or not in 2022 according to your budget?
So we are talking about the higher EBITDA of start-ups in 2022?
So this is a fair assumption?
Next question comes from Mr. Jerry Dellis from Jefferies.
First question has to do with your fixed broadband ARPUs, your retail ARPUs, please. It looks like from your disclosure that retail broadband ARPUs are declining about 7.5% year-on-year in the third quarter. And that's obviously interesting in a situation where your fixed line churn is so low. Would the correct interpretation of that be there's a lot of retention investment going on, perhaps repricing down the back book in order to keep those customers loyal. If that wasn't the case, then with such a stable churn situation, why would the broadband ARPU be under so much pressure? And then my second question, if I may, is just to sort of understand a little bit more about the data analytics work that you said that you've been doing, which supports your strong view that it's possible to take up prices in this segment. I'd be interested in the sort of the segmental work that you've done that sort of underpins your confidence in that and your ability to sort of base the revenue guidance upon it, please?
Okay. With regards to your question about data, that's a long question that if you'd like, we'll probably follow up offline. But basically, what we have done is that over the last couple of years, we have created a major data lake with all the information about our customers, and we have a group that -- actually I see our Chief Data Officer that is very eager to explain you what he's doing. Okay. I didn't want to -- he's a kind of, I call him, crazy scientist, but I let you speak, Maximo, at your risk and peril. You made a question, now you get Massimo Arciulo, our Chief Data Officer.
Thanks a lot. Maybe I'm only crazy, thankful to be appointed as a scientist. In a couple of minutes, I'll just say that what we are doing in terms of the office. We are building what we call data market automation. That is basically to perform knowledge discovery through inference that are statistics that address pattern among the client behavior. We do it through -- we are doing it to our built in-house engines. The first one is a machine learning engine with every call to client, that generally, it shifts all the information to entities that are flowing into the data river, and then we get the information that are generally hidden. After having done it, we immediately pass and process those information into their marketing field. So generally, the discovery knowledge is [indiscernible] TIM, we are doing rather than being segregated in artificial intelligence here that we are generally processing them, orienting them through the execution on the operational side. And so that's why my department is strictly working with staff of [indiscernible] in a way we can consider the data market automation as an integration -- an integrated machine. So this is currently the main point. So I'm going to conclude the data market automation, discovery knowledge, patterns -- hidden patterns among the customer profiling. Finally, it's the actions performed to [indiscernible] to reduce job.
I don't know if it's clear now. But I'm sure it is. But if you'd like to follow up, and if you're interested in the field, we are happy to follow up with a more detailed session dedicated to. Or if somebody else is also interested in data analytics, on the sector and would like to participate other than competitor, obviously, please ask Carola and we'll be happy to organize a demonstration. I think your second question on the ARPU. And ARPU in fact, is improving its year-on-year performance. What's paradoxically, and I think I mentioned that in the past as well, paradoxically, the reduction in churn, which is a very positive thing is as a short-term negative impact because we don't do a lot of activation. Basically, we do not have activation fees which are part of the ARPU. And so in the short term, they tend to penalize that as part of -- as an impact on the numbers. But we do not reprice the back book because being the market leader we are prevented from our legal authorities to do retention. And in fact, this is particularly exciting, our churn and our market share growth because we are not allowed to do retention while, of course, we do know that other players do that. But this has been all done on our own. And in fact, one of the reasons that you see CSI improving and Net Promoter -- NPS also improving. So fortunately, and it's on Page 16 of our presentation, our statistics there and KPIs are quite satisfactory. And in fact, this is a sort of schizophrenic presentation because on the one end, we had an unsatisfactory performance on the P side of the equation and, basically, the prices -- the market prices brought us to a lower result. On the other end, on the Q side, on the market base and so on, all the KPI were quite positive. In fact, they have been the most positive since we started these conversations. I would ask -- I think we have time, Carola, tell me, for one last question?
And indeed, the last question, please.
Our last question comes from Mr. Keval Khiroya from Deutsche Bank.
I've just got one question. You showed the primary network within the strategic initiatives update chart and you referred to it as well. When do we think about possible actions and then to how much you can say, but would the most likely scenario be for that primary network to be potentially fold into FiberCop perhaps as part of any single network deal? Or are there other structures which could work for this type of primary network assets?
I hate to invoke the Fifth Amendment, but this is a type of question that it's basically -- I'll have to answer to the Board first, and then I can do to the market. But as I was saying at the beginning, basically, if we had to take the single biggest chunk of value in TIM, as you immediately realized, it's there. So we'll have to revisit our option. But I think that there is a general trend in the industry. So let me speak more than just about TIM, about having some time technical and services separated. I know there is a number of operators that are looking at that. Some have already done it. Sometimes it helps you clear. Remember, you still have to always remember that this is an industrial group, so we should not lose economy of scope, which are quite important. But sometimes having a clearer picture allow you to identify the area of value and the area of cost and the area which you want to intervene. So clearly, in this exercise that I'm about to be starting this part of our balance sheet, it's one in which it's going to be a lot of focus. And I'm sure that we'll have other opportunity to speak more at length on this. With that, I think I would like to pass the mic over to Carola again. On my side, thank you very much for your attention. Looking forward to talk to you next year with our business plan.
Thanks a lot from my side as well and from the whole team, and we're available, as usual, to answer any question you may have. Thank you. Bye.
Ladies and gentlemen, the conference is now over. Thank you for calling.