VEON Ltd (VEON) 2017 Q4 法說會逐字稿

Okay. Good afternoon, ladies and gentlemen, and welcome to our London office. I'd also like to say welcome to those joining on the webcast and the audio call. Before we start with the agenda, I just need to read out something from the disclaimer that's on the screen. Forward-looking statements made during today's presentation involve certain risks and uncertainties. These statements relate in part to the company's anticipated performance and guidance for 2018, future market developments and trends, operational network development and network investment and the company's ability to realize its targets and strategic initiatives. Certain factors may cause actual results to differ materially from those in these forward-looking statements, including the risks detailed in the company's annual report on Form 20-F and other public filings. The earnings release and earnings presentation, each of which includes a reconciliation of non-IFRS financial measures presented today can be downloaded from our website.

Right, let me now run through the agenda and the speakers today. We will start with Jean-Yves Charlier, our CEO, who is going to go through the 2017 achievements and our 2018 strategic priorities. We will then have Trond Westlie, our CFO run through the 2017 financial results and the 2018 targets. There will be a break of 20 minutes. After which Kjell Johnsen, Head of Major Markets will give an update on the situation in Russia; followed by Jeffrey Hedberg, the CEO of Wind Tre, who will give an update on the Italian joint venture. Jean-Yves will then come back and make a few final remarks. There will be an opportunity for people here to have refreshments and further discussions with the presenters after the presentations.

And with that, I'd like to hand over to Jean-Yves.

Thank you, Richard. I just need the clicker. Perfect. Good afternoon, everyone, and welcome to VEON's 2017 investor presentation. My apologies first and foremost for my voice. Three days ago, I had no voice left. So I'm going to try to make it through this presentation. In fact, earlier today, I was looking up for some quotes around losing your voice, but it seems that there are no quotes for the matter. People don't have words left. So I'll try to make through this presentation.

I think we've really got an interesting agenda for you today. I'll be covering, as Richard said, not only the highlights of our financial results as well as obviously progress on our strategic initiatives, but we'll also have deep dives, obviously in these financial results, but deep dives in Russia and Italy, which contribute to a very significant part of VEON's value today, but also do a deep dive on the progress of our digital strategy.

Overall, 2017 has been a good year for VEON once again. I think we've delivered on most of the metrics that we set out to grow. We saw solid top line growth at 1.9% organically, 6.6% in actual terms. In fact, in 2017, we really had some tailwinds from a currency point of view, and total revenues reached $9.5 billion. That was really fueled by very significant growth in our data revenues across the board. They grew organically by 25.7%, in the natural terms by over 30%.

We reported EBITDA growth of 11% to reach $3.5 billion. That was a 7.5% organic growth rate. Our EBITDA margins, underlying EBITDA margins were slightly below what we had guided to -- for to a tune of about 0.9 percentage points, mainly due to the underperformance in Algeria and Bangladesh. We'll do some deep dives on those 2 specific countries and the underperformance that we saw in Russia during the fourth quarter.

We continue to deliver on our CapEx guidance. As you recall, we had guided and took a number of initiatives to really drive CapEx from roughly 21% to 15% over the medium term. We delivered during the course of 2017 CapEx number of 15.4%. What's very interesting about this is that we had as many upgrades in our network in 2017 than in 2015 when we were spending 21% of revenues, or CapEx to revenues. So we've done this much more efficiently. We've done this by centralizing purchasing, and we've done this by moving to both Huawei and ZTE as main providers of our network technology.

I think what we're particularly proud of, obviously is the very strong underlying equity free cash flow number that we've delivered. Close to $1.1 billion, well above the guidance that we had provided. And that's really been the driver of this whole transformation strategy, to really drive cash flow to generate shareholder value. On the back of that, we are pleased to announce today that the Supervisory Board yesterday declared final dividends of $0.17 a share, bringing the total dividend for the year 2017 to $0.28 a share. That's a 22% increase in dividends for shareholders. And I think that is very much in line with our dividend policy of having a progressive and sustainable dividend policy for our shareholders.

In 2015, we set out a 3-year strategy really to, on one hand revitalize VEON's business across the world, and I think we're really delivering on that agenda today, but we also set out a medium- to long-term agenda to reinvent this business. Because ultimately at the end of the day, we feel that we need to find new levers of growth. We need to find new levers to be able to make our cost structure more efficient. We need to find new levers to engage with the 240 million consumers or so that we serve across the world. And obviously, this is very much at the heart of a medium- to a long-term strategy, but we feel that we're starting to make progress and I'll deep dive into that later today.

At the heart of both revitalizing and reinventing VEON is really this key objective of substantially increasing equity cash flow for our shareholders to be able to obviously provide the appropriate dividend policy and appropriate dividend levels that may get meaningful for our investors and shareholders to own VEON stock today.

During the course of the year, we've really seen an improvement in all our key metrics. And this has now been 2 years in a row that we've been delivering on all these dimensions. And as I said, the elements where we're particularly proud is obviously the 31%-or-so growth in our underlying equity free cash flow, which has allowed us and allowed our board to declare a dividend of $0.28 a share for 2017, an increase of 22%-or-so.

During the course of the year, I think we've accomplished a lot. Once again, we've really delivered on these strategic initiatives, particularly, in terms of world-class operations. We said we would increase the free flow 3 years ago, and I think we've delivered on substantially increasing the free flow, which is just shy today of 30%. We said we would improve our capital structure. And during the course of the year, we really have improved our capital structure, and in Q4 I think we undertook a very successful refinancing of Wind Tre in Italy, delivering a very significant savings in terms of our interest cost on the joint venture.

We continued during the course of 2017 to fundamentally improve corporate governance with the appointment of Ursula Burns as our Chairman, and today we have more unaffiliated directors on the board than ever before. And during that time, we've continued to improve and strengthen the control and compliance environment, which had been a fundamental weakness of the former VimpelCom for some time. We also during the course of 2017, really delivered on consolidating our portfolio and moving progressively more towards of an asset-light type of telecommunication model.

We did the GTH share buyback. We announced some mandatory tender offer for GTH just before Christmas. That's taking longer than what we originally expected. I'll come back for the reasons for that. We agreed to dispose of our tower portfolio in Pakistan for over $900 million. And we expect to close on that transaction in the near future. Just before Christmas, we unlocked the Uzbekistan situation. We were able to upstream $200 million of cash back to headquarters, and I think really unlocking complex structural issues that the group faced. In fact, Trond will talk about this, but I think we've make substantial progress in the last 12 months in ensuring that every one of our OpCos is contributing to a dividend flow back to Amsterdam to support the dividend back to our shareholders.

And finally, we've continued on our journey to dispose of nonstrategic assets with the disposal of Laos during the course of the year. We've really also been focusing on accelerating the growth opportunities in our portfolio. Mobile data is growing very fast, close to 26% year-over-year. But in other areas that we highlighted to you in the past, our B2B business is growing faster than our B2C mobile business. It grew during the course of the year close to 4%. Our FMC strategy across the world revitalizing the fiber assets that VEON has is also gaining traction, 265% growth there. And in certain markets, we are really seeing significant traction such as in Ukraine and Kazakhstan, and Uzbekistan and in Russia. And Kjell will talk more about the Russian opportunity. We made a bold move in Russia during the course of 2017 by announcing the split of Euroset. And right now, we're in the course of really owning the destiny of our distribution strategy in Russia, which we believe over the medium to long term will make a very significant impact in terms of our business.

We've recently strengthened our spectrum portfolio in Ukraine and Pakistan and Bangladesh, and the strengthening of our spectrum portfolio in Bangladesh has been very critical for us as it has been a structural issue that we faced against the other 2 main competitors in the marketplace. And we are very pleased with the results of that spectrum auction in the last few days whereby we strengthened both our 3G and 4G portfolio. When we look at this strategic initiative, really of accelerating our growth rates, key to this agenda now is really unlocking the issues in Algeria and Bangladesh. This is what we've got to get right during the course of 2018 be able to deliver better top line growth for the group.

And we've continued transforming our cost base. I think we've been immensely successful in terms of taking to the appropriate levels our CapEx spending and investments. And many of the detractors are saying easily we're just cutting for the sake of cutting. As I said, during the course of 2017, we upgraded more sites in our network when we're spending 15.4% of CapEx than we did in 2015 when we were spending 21%. And it's all down to smarter investments. It's all down to having centralized our purchasing. It's all down to transferring our network-preferred partners to Huawei and ZTE that are allowing us to have much more efficient price points in terms of our investments.

We also have reduced our cost base internally. That's allowed us really to keep it relatively flat, while propelling our investments in digital. And we recognize to date that just keeping our cost base flat going forward is not going to be good enough. We've got to be able to invest in the future, invest in digital, but at the same time, we've got to be able to take our cost base down year-over-year. And that's one of the key objectives that we are setting for ourselves going forward.

But we've made a lot of changes here, right, and investments such as purchasing and consolidating purchasing on a worldwide basis. In 2015, less than 20% of our purchasing was done on a centralized basis. Today over 80% of our purchasing is done on a centralized basis, allowing us for very significant efficiencies. We're right now in the process of consolidating all our roaming wholesale type services across the world to drive exactly the same type of efficiencies. There are multiple initiatives of this type within the VEON group to really drive a much more efficient cost structure than what we've had in the past.

Let me now turn to giving you an oversight of our portfolio and what we are doing well, and what are the areas that we need to focus on. Obviously, we have quite a number of our OpCos that are performing exceptionally well. I certainly would put in that category, Ukraine, Pakistan, Kazakhstan and Uzbekistan. Right? These are businesses that have very solid positions in their marketplace, and have a historical track record of very, very solid performance, and we believe that we can fundamentally continue those type of trajectories even in countries such as Uzbekistan, where we have pressures from a tax point of view that obviously weigh on our results or currency devaluations, which we can't do much about, that obviously weigh in terms of our actual dollar-reported results.

I think what we've seen during the course of 2017 is a much better performance overall in Russia in spite of the underperformance in Q4. Kjell will talk about that. What we've got to get right is continue that trajectory. We've got to get right our fixed-line business in Russia. And we've got to make it a success of the Euroset monobrand integration, which really we feel can make a difference over the medium to long term.

The areas that really need our focus and our attention. First and foremost, Italy. We have to defend our position in Italy. Right? And we've got to continue delivering on our synergies. We are very committed to delivering our synergies. And as Jeffrey will talk about, I think we've been doing a very good job from that point of view in taking cost out of the organization. What we haven't been doing so well is in terms of the top line. And we've seen a very competitive Italian marketplace over the course of the last 12 months, probably more competitive than what we had expected. And we are putting in place the appropriate initiatives to ensure that we defend our positions with the fourth entrant coming into the marketplace later this year.

The black spots really have been in our portfolio, Algeria, because of the macroeconomic issues as well as the structural issues, and Bangladesh in terms of structural issues. We believe that 2018 could be a turning point for those 2 businesses. I've always said that turning around these 2 businesses would be a medium-term type of turnaround, there was no quick fix in these 2 countries. But we feel that as you will see in a minute, we have fundamentally fixed the basics in both those countries. And we hope to be able to start delivering better results more in the short to medium term than sort of the medium, longer term of what we had said a year or 2 ago.

Let me maybe do 2 deep dives at this point, both on Algeria and on Bangladesh to give you a perspective of what we've been doing. Two big issues for us in Algeria. Right? First big issue has been the regulatory environments. Right? A government that was against this asset some 3 years ago. It took a long time to resolve these sort of structural issues and create a real partnership that we have today with the Algerian government. We've also had to deal with macroeconomic environment that is weak and perhaps weaker than any of our major assets that we have across the world.

We believe today that we've addressed most of the regulatory issues. The last one that we need to address is in terms of MTR symmetry. We've made progress during the course, latter part of 2017, but we are not in full symmetry today, and more progress needs to be made in terms of 2018 to achieve that. But also we need to work with the regulator and the government to ensure that there is a moderate regulatory environment, not just an acceptable one in Algeria. A modern environment that allows for network sharing, a modern environment that allows us to pursue our asset-light strategy and to dispose of our towers there in that environment.

We had to do a lot to fix the network in Algeria. This was a network where we couldn't invest for years. Right? And we believe that we now have the best-performing network in Algeria. We cover 75% of the population from a 3G point of view and 25% of the population from a 4G perspective. And all the recent studies indicate in Algeria that the Djezzy network is by far the best network today in the Algerian environment.

We also needed to address our commercial performance. Our offers were too complex. They were old. I think we've restored completely new sets of offers, of tariffs, of products in the marketplace. We are revamping our monobrand distribution. In the course of the last 12 months we've opened 52 new flagship stores in Algeria. And ultimately, we've had a strong focus on the new growth opportunities such as B2B where we've been growing in Algeria by 7% year-over-year.

We believe today that we have a very strong management team in Algeria, headed up by Matthieu Galvani. All the key positions to date are filled. And we've really been working on making this organization much more agile overall. In the last 2 years, we've taken out of the organization 6 management layers to make this business much more agile. And from a cost point of view, we've reduced the workforce at the same time by 25%. These are substantial moves to reposition this asset so that it's competitive again in this Algerian marketplace. And once we've done on this and in spite of the competitive pressures, you've seen that we've held up in Algeria some pretty healthy margins, around 47.5% for the year, but lower in Q4, but we've really been, whilst under attack and whilst we're restructuring, really taking cost out very effectively out of this business.

We're also beyond the traditional telco side, really starting to invest in the future in creating a digital future for Djezzy. We implemented in the course of 2017 the group-wide DMP platform in Algeria. We're going to deploy the new DBSS Ericsson system during the course of 2018. And our self-care app and our move to simplifying and enhancing our customer journeys has been the #1 downloaded app recently on the Google Play store in Algeria. And these are, I think the first green shoots of what we want to make of this business for the future.

Very much of the same we've been doing in Bangladesh. Right? We've had to address a regulatory environment that's not been the most modern. And there is still work obviously to be done there, I think in 2 areas. Certainly to be able once again to do network sharing-type of arrangements and disposing of our towers so we move towards a more of an asset-light model, but we fundamentally believe also that in this 3-player market, there are SMP issues that need to fundamentally be addressed by the Bangladeshi regulator and government, and that's part of our agenda for 2018.

Structurally, we had a week network and a weak spectrum portfolio. During the course of the year, we've addressed both. We believe today that we've got a network that will perform well, and will perform well particularly during the monsoon season where there are electrical shortages across the country. We've made substantial investments in upgrading the network, substantial investments in terms of batteries and alternative power sources so that the network is up and running during the forthcoming monsoon season.

But more importantly, we had a spectrum deficiency, which we've addressed last week by investing $345 million and creating a spectrum portfolio that is today competitive for 3G and 4G. And in fact, it is the most competitive spectrum portfolio on a spectrum per customer basis today in Bangladesh. So we believe that we've ticked all the boxes there for the business to perform once again.

We've been really repositioning Banglalink as a digital attacker brand in the marketplace. We've recently launched new simplified offers. We are also, like in Algeria, revamping distribution. We've just in the last few months increased points of sales by 50% in Bangladesh. We also believe, like in Algeria, we have a strong management team in place that is led by Erik Aas and once again all the key management positions in that marketplace are in place. And we've done the same thing. We've cut management layers. We've streamlined the organization. We have taken out close to 50% of the workforce in the last 2 years in Bangladesh to make that organization streamlined to be able to continue maintaining reasonable margins. And you see during the course of the year, they were just north of 40% despite of being slightly down in Q4 as we made these investments.

We are doing the same thing in terms of our digital agenda. We're really putting the green shoots for the future. We deployed our DMP platform during the course of 2017. We will launch our Ericsson digital BSS platform during the course of 2018. And again, we are making the same investments to digitalize the core, to digitalize our customer journeys, to create engagement with our customers with new self-care and digital platforms.

Let me now turn to where we are on the GTH mandatory tender offer. As we announced in December, we believe that this is an important strategic milestone for the VEON group to be able to directly own these assets in Pakistan, in Bangladesh and in Algeria. This tender offer remains subject to approval by the appropriate Egyptian authorities. We believe that we have taken all actions required for such approval, and that the matter is currently held up by the authorities in conjunction with some disputed GTH tax issues.

We've been working with the authorities. And our desire is to find a path forward with the Egyptian authorities to basically unlock this mandatory tender offer. But at the same time, we are prepared to consider all other options that allow us to achieve this long-term objective of owning more of these assets for the future of VEON. There's probably not a lot more that we can say today, and we are working closely with the authorities to unlock as soon as possible this set of matters.

Let me now turn to our digital agenda. And what I'd like to focus on today is really how we are driving a profound set of initiatives to really reinvent VEON. And as you know, this is no easy matter to reinvent a telecom operator. Right? I've always said that this was a long-term strategy. And during the course of that strategy, we'd have a few setbacks. But that ultimately our vision is very clear. It's a vision where we have to reengage with consumer. It's a vision where we have to offer new services beyond connectivity with consumers. It's a vision where journeys need to be digital. It's also a vision about reinvest -- reinventing our own business model, moving away from a brick-and-mortar model to much more of a digital model, that allows us to continue to streamline our own business, that allows us to take more cost out of the organization than what we've been doing up to now in the traditional manner.

So today, we are basically focused not only on transforming our business model, moving it away from brick-and-mortar, but we are also very much focused on creating the new services for the future. And what we're doing in the course of 2018 is making these 2 initiatives very separate from one another, so we can really boost these 2 programs within the organization. And within that, we have 5 supporting initiatives. Obviously, the VEON platform remains key to this strategy and delivering on this vision of engaging with consumers in a different way and creating new revenue streams.

But our digital transformation is much more than just the VEON platform. And I want to make that clear for all investors and analysts today. We are obviously very much focused on creating a big data layer within our organization so that we can personalize our offers, whether they are new digital services or our traditional connectivity offers. We want to be able to operate like an Internet company, and have all our data collected in such a way that we can personalize these type of services. But in order to do that, and that's our third key initiative, you can't do that with the traditional legacy BSS systems of operators.

We have over 300 different systems within VEON today. Some of the systems are stovepipe. Some of these systems are not real-time. No one can create a digital strategy without fundamentally changing the BSS architecture of the telco. Right? And this is complex. Right? Very few people have been able to fundamentally change profoundly the BSS infrastructure on a global scale.

We're also, fourth initiative, doing the back-office. Right? We are changing our ERP systems, going to cloud-based systems to have the agility from an end-to-end point of view and we're virtualizing our networks. Because fundamentally we are prepared for 5g, we believe that there needs to be a split between the hardware dimensions of our network and the software dimensions of our network. And I believe that VEON today is leading the industry. I will touch on a few of these strategic initiatives today to ensure that everyone has a better understanding of what we're doing.

The VEON platform, we are more than ever committed to its success. And this is a medium to long-term strategy, creating a platform that can compete in the marketplace against the best of the Internet platforms. And during the course of 2017, we launched the VEON platform in a number of countries. We have seen a number of green shoots. But we've seen that we got to do a lot more to make it immensely successful. And one of the key elements that we are announcing today is that the platform was specifically focused on the elements of messaging and marketplace. Right? And that we are going to take out of the platform all the self-care components and put that within our whole strategy to focus on digitalizing the core. That ensures that customers better understand the value proposition of both components and both platforms that we are offering on the marketplace.

The second element that is absolutely critical then for us is that we are accelerating the marketplace dimensions. We were expecting to launch marketplace 2 to 3 years out. We are now expecting to trial marketplace during the second half of 2018 because we've seen real interest from both global brands and local brands to really be part of such a platform in the markets where we operate today.

During the last few months, we've had over 8 million downloads of the platform. We today have more than 3 million monthly active users, and we've seen a number of green shoots. In Pakistan, the app was rated #1 for a number of weeks in the App Store. In less than 90 days, we had close to 3 million downloads in Pakistan alone. In Georgia, what we're seeing is that the messaging element is what consumers like. In Pakistan and Russia, the in-app offers and the content element is what is being core to the platform.

So the further we move on, the more we'll start reporting on some of the key metrics of the platform, but this gives you to date some of the highlights of both the strategy and the green shoots that we've seen in the marketplace.

What we've also seen is the interest and the momentum that the platform is gathering in terms of partnerships and brands that want to promote their products and their content on this platform. During the course of the year, from a start of 0, we signed over 200 partnerships for the platform, 120 content partnerships, 97 offer partnerships, 28 payment partnerships. And today, we're working in the first quarter on over 100 new partnerships to add on to the platform. So a lot of interest, a lot of momentum, and hence why we are absolutely positioning this as a WeChat-like in the markets that we are operating today.

Let me turn to how we are going about digitalizing our core business. It really has at the heart 3 main components, everything that's customer facing. We believe that at the end of the day all customer touch points should be digital. There should be no need for any customer to have to walk in a store or call someone to get support. Everything should be done online. Everything should be done at the touch of a button on an app within 2 or 3 clicks. That's our vision. And that is what we are building today from our customers alongside the VEON platform.

But in order to do that, we've got to really fundamentally rethink our networks, rethink our IT, rethink our IT architecture to be able to deliver both for our self-care app and the VEON platform, personalize offers, personalize services, but more importantly, on a real-time basis, right, with no lag. Customers in this world are not ready to wait 30 seconds to get a response on a click of a button. Everything has to be instant and easy and simplified and transparent.

But we believe fundamentally at the same time, we've got to reengineer completely our administrative systems and processes within the company. So the customer-facing side, it's about taking a clean sheet and really rethinking all our customer journeys and making them end-to-end digital. It's about offering new channels, improved self-serve. And as I said, real-time response. On our networks, it's making them ready for more data growth. Making them ready for 5G. Taking the cost out fundamentally of our network and revamping our systems to offer these new services, but revamping our systems at the same time to take cost out of the organization.

And all these things flow together. This is not about recreating a stovepipe architecture as all the telcos have done in the past. We have to have data flow, real-time data flows between our networks and our BSS systems, our ERP systems and then ultimately our data mining systems to offer real-time thousands, millions of different offers on a personalized basis every day of the year for our customers. So we're very much focused on this integrated strategy.

DMP is absolutely core. What's very new for a telco operator is to be thinking big data. Currently, and you see this type of new office, this is a type of standard that we're rolling out across the world. Because we are recruiting software engineers, we're recruiting data scientists. Right? And so having modern offices like these where nobody has an office, where the technology is the best, where people work on some of the most compelling projects is what is at the core of our transformation. We believe having a DMP platform is essential, as I said, to personalize our services, to understand every one of our customer's requirements. Not just doing what the industry did in the past, to send an SMS to tens of millions of customers, and to ultimately have very low return rates. Those models are absolutely models of the past.

But we believe DMP needs to be used on how we upgrade our networks, how we invest in our own organization. And ultimately, we believe that at the end of the day, this should enable us to probably drive a 1% EBITDA impact over the medium to long term in our business by allowing us to decrease our sales cost, our marketing cost and make more efficient investments in our networks.

Obviously, the bigger piece of all this is whether or not we can succeed in changing the 300-or-so stovepipe systems that are not in real-time but VEON has across world. And we believe that this strategy is absolutely essential to not only offer real-time data, and impressive data that all telcos have around the world, but also leading to a substantial impact on our cost structure. Our objective is to take our BSS costs and to halve them basically as a result of this strategy. We believe that we can take our BSS costs from roughly 2% today of revenue to 1%. Right? And at the same time we believe that we can fundamentally improve customer engagement, fundamentally improve customer service, customer touch points at the end of the day.

And we've partnered as you know with Ericsson. And we are very pleased to stand today here in this presentation to announce that we've implemented the first country with this new system stack on time within spec, and that's Georgia. And we have a large number of countries to go during the course of 2018, Algeria, Ukraine, Bangladesh, and we expect with Ericsson to have completed the whole spectrum in 2019. We are doing exactly the same thing in Italy. Jeffrey will touch about -- talk about this so is Kjell in Russia with very aggressive agendas in terms of really transforming our architectures.

So what are some of the first results in Georgia, to give you a perspective there. We've gone in Georgia from 27 systems, all stovepipe, multiple vendors, one integrated system. Right? We've cut 49% of our tariffs and our products in doing so to simplify completely our offerings to our customers. Right? We are now real-time. We can launch an offer of any type in a few hours or less than an hour in Georgia. It took us weeks to prepare an offer in the past. Right? So this is the platform to start operating like an Internet company ultimately. And it offers us the opportunity to provide our customers with much better self-care tools, digital touch points along every one of the journeys and a greatly improved customer satisfaction, we believe, at the end of the day.

But we're also achieving at the same time the cost cutting as a result. Our BSS stack in Georgia cost us 1.9% of revenue. Today, our new stack is 1% of revenue. Massive savings ultimately as a result of this strategy. We're doing the same thing, ultimately in terms of our networks, we want to decouple hardware and software. The industry is at -- just at the beginning of this. We've already done this on most of our backbones across the world. It's a very important long-term strategy because we believe again that it will not only prepare us for better services, it prepares our networks for 5G, but ultimately, it also drives much more efficient cost structures in our business. To put it in perspective, we recently launched a vEPC tender, which is a Virtualized Evolved Packed Core Network backbone tender. We reduced by a factor of 5 the TCO compared to the legacy architecture that we had in place. So this isn't about 5% to 10%, 20% cost savings. The magnitude of cost savings of virtualizing the network and modernizing the network are absolutely significant, we believe.

So I think this gives you an overall first perspective on the results, where we see our business going, some of the deep dives of things that we're working on. I think we're particularly proud of continuing to accelerate on this digital transformation. Because we think it matters for the future, and hence the investments that we're making here.

I'd like to end my presentation really on the 6 key priorities that we as a management team have set for 2018 to give you perspective of what is our agenda. First and foremost, this is a year where we have to start delivering on the transformation of Algeria and Bangladesh. And I don't mean delivering on fixing the problems. We've been fixing the problems. We have to start seeing the results, right?

And I think that in the second half of the year, I hope that we'll start seeing the type of results that show improvement, but as a management team we are very committed to fixing those 2 dimensions in our portfolio.

The second one is defending Italy. Although Italy is not consolidated, we believe it's a big driver in terms of our potential growth of valuation. And it's absolutely critical for us to defend our positions in the marketplace with a new entrant, but perhaps equally as important is to drive the synergies and perhaps even deeper cuts than what we had originally envisioned around this business.

We've got to continue performing in Russia and Euroset integration as well as fixed-line are core to that agenda. Kjell will address that.

Fourth, real priorities that we've got to do more in our cost structure. And a number of investors have challenged me and Kjell and Trond and our wider management team about our corporate costs. And over the course of the last 2 years, we've had to increase those corporate costs really to get VEON performing better, performing at world class. We've had to consolidate a number of functions. 2018 is a year where we feel we can start taking our corporate costs down. And as I said, our objective ultimately is to have an agile operating model within the organization, and not just keep our cost structure flat throughout the whole group, but attempt to take it down as we invest in the future.

The fifth strategic priority is obviously to continue accelerating our digital agenda. We think the vision is right. There will be a number of setbacks along the way. But we want to continue driving that agenda because we believe it can fundamentally deliver value for our shareholders.

And the sixth key priority is really to unlock the MTO in Egypt. We think this is a critical value driver for our shareholders and our investors. And as a management team, we are very committed to doing whatever it takes to unlock the mandatory tender offer of GTH.

Now the underpinning of all this is ultimately to continue to grow cash flow. That remains -- may be our ultimate #1 objective. And Trond will take you through guidance. We expect to grow substantially, once again, cash flows during the course of 2018, because we want to fundamentally continue to underpin that progressive and sustainable dividend policy for our investors.

Thank you very much for your attention. I'll now hand over to Trond who will take you through the numbers in much more detail. And whilst I lost my voice at the beginning of the week, but had no flu, unfortunately, Trond had flu over the weekend and beginning of the week. So like me, he might not be in the best of shape, but we're here to deliver this presentation. Trond?

Very much so. Thank you, Jean-Yves, and good afternoon to all of you as well, from me. So I'm going to take you through the -- some of the financial results and the targets for the year. And I will first start to give you a few reflections from my point of view, from -- on VEON coming in at the end of last year, I have the -- continue to have a view. And to give you some reflections.

I do think that VEON has come far in the restructuring of the company itself. I do think that the only real outstanding item on the restructuring part is the GTH. As I mentioned in bullet #3 here. And from there on, the corporate structuring is really a simple situation. And then when it comes to the capital structure, I will go through that in a bit. I really think that become a far way in actually doing a lot of things on the corporate structure part and the corporate structure on the capital. We have the guarantee structure and refinancing elements. So all that has come far.

Coming to the governance control and compliance. Jean-Yves mentioned that with the new chairman having set up of -- in the independent board, making sure that we have the process, procedures, control elements in place, and the compliance as a result of focusing on that for the last few years.

I think we have confirmed that. We've been Sarbanes-Oxley compliant for many years. So the controls and financial reporting has been there for quite some time. So all in all, I actually think VEON is in reasonable good shape there as well.

Cash flow has improved significantly. We have -- I think VEON had managed to make the statements [within] a [peg] around the cash flow relative to the adjustments and so forth. I've tried to go far and clean that up a bit and make sure that we can go straight from the numbers that you actually can report and see. So I think I'll take you through that, but even there you actually see the development and improvement year-over-year of coming into a more normalized company in the cash flow area.

And the last point is, of course, the focus on the general and corporate cost. I do think that as Jean-Yves said, some of those investments in headquarter costs has been needed to actually come to quality, to process, to have the company be as it are -- where it is today. But having said that, I do think that we cannot have that level, so we need to come down. I will address that as well.

So I do think that, my coming -- my mindset in coming into this now is really that we are in a mature industry. We're in the middle of a value chain. We have to recognize that and as sort of a telco industry with a digital agenda. That digital agenda is going to be important for our future and capitalizing on the future, where we actually see the future. But we also have to drive the telco part on the continuous improvements, on underlying improvements going forward. So that's my overriding mindset of going into this.

And then going a bit more into details on the numbers. This is the 2017 full year numbers. The reported numbers of '16 were just short of $8.9 billion, going up to the short of $9.5 billion. So in the reported numbers, we are going up with 6.6% this year. Voice coming down. The 155 you see to the left of the top, you actually see that it’s the Warid first half year that we did not consolidate. So it's just to show that year-over-year, the 6.6% is also taking the benefit of Warid having a 12-month consolidation and not the 6 months consolidation.

Voice revenue coming down and Data and Mobile Financial Services coming up. Data -- so that sort of driving the overall together with the Forex that leaves us to a 6.6% growth.

If you then to take that and decompose it down to the companies through the business units, you see that Russia, Pakistan, Ukraine, Uzbekistan is developing -- is delivering the growth aspects underlying in the organic terms, and then as Jean-Yves had already mentioned, Algeria and Bangladesh is the one that is challenging the points that we will addressing during 2018, and hopefully, we're going to see a turnaround.

If we then see the EBITDA evolution. I do think that the growth in EBITDA coming from the Forex $340 million. It really is 3 buckets of -- the 6 companies divided in 3 buckets. It really is Ukraine and Pakistan delivering the growth aspects that's driving the pickup. It's Russia and Uzbekistan basically being flattish to slightly negative. And it's Algeria and Bangladesh taking the numbers down in evolution. So that's really the trend line that we're looking at.

If you then look at the year-over-year trend line, if you go from the left on Russia, I'm not going to say very much because Kjell is going to cover that later, but we see a slow growth year-over-year in fourth quarter in organic terms, and then a decline in the EBITDA level. And there is a few elements going into that. It's a slightly weaker trend line, some cost pick up and launch of VEON. But as I said, Kjell is coming more into that.

Pakistan is doing well. 4.2% growth year-over-year. So a good growth. And you also see that the synergy effect and the effect of having a combined network is also driving itself into the results as you can see on having an EBITDA increase of almost -- short of 40%.

Algeria and Bangladesh, coming down 10% year-over-year. But as you can see, we are better at compensating on the cost efficiency side in Bangladesh, even though Bangladesh is coming even more down as a result of the network cost and the structural OpEx that we have in the fourth quarter as a result of the outage in the network as we had -- of the cost coming in.

Ukraine, market leader, doing very well, 10% growth in revenue year-over-year, and also an increasing EBITDA. And the EBITDA is increasing dramatically because it has a 40% increase in there so coming up to a high 50% margin, which is really a very good result.

And then Uzbekistan. Good revenue top line organically. The tariffs there have been linked to U.S. dollars for the first 9 months. So as a result of that, of course, it's trendline, but in the fourth quarter sort of flattening out as a result of liberalization and tariff being fixed. On the EBITDA and the cost side, a lot of taxes have come in and increased the tax base. So a lot of the increase in taxes or cost side in Uzbekistan is really non-controllable. On the controllable cost side, we have done a good job and have actually year-over-year improvements on the controllable cost side level in Uzbekistan.

When it comes to the corporate costs, the corporate costs in 2017, which is then consisting of Amsterdam, London, costs for digital and also cost for services and projects, specifically M&A, legal costs, the Enterprise Service System, the ERP system, global procurement and so forth, we see that the -- that gross cost is actually getting up with approximately 2.5%. So not growing much year-over-year from 2016 to '17. Having said that, as both Jean-Yves and I have been saying, this cost has been needed to grow. And remember, year-over-year, from '16 to '17, we have also upped the digital part of this cost. So other cost has been coming down, but in total our ambition is actually to take down this cost with 10% and 20% year-over-year, next year -- or this year, sorry.

Going now into the quarterly numbers. We have a revenue number of $2.3 billion, down from 1.4%. Most or all of that is really relating to the Uzbek currency element, because the underlying organic element is a slightly growth. And then you see the service revenue of $2.214 billion, leaving us at $753 million in EBITDA. The EBITDA is a decline, a reported decline of almost 4%. As I said here, Uzbekistan currency is actually giving us a 6.6% decline as a result of the reported currency there. Underlying, the EBITDA has a small organic growth.

Coming to the net financial income and expenses. The same level as last year, except for the option, the put option that we have in the Warid transaction and that has a pricing element that needs to go through the P&L, and that creates P&L effect of $38 million. Forex and other gains, it's the Forex losses of HQ and Russia mostly, plus the $49 million of charges we actually took from taking $200 million out of Uzbekistan during fourth quarter.

The $156 million of joint ventures is Italy. And that leaves us with a loss before tax of $285 million, and from continuing operation of $378 million. Just to be finished, when we're finishing on the technical numbers. I'll finish on a technical elements of (inaudible) IFRS 9 and IFRS 15 will be implemented beginning of '18. IFRS 9 is going to do us a $20 million pretax charge in the retained earnings, and IFRS 15 is actually going to give us a pre -- the one-off gain in retained earnings at $90 million -- at $95 million, sorry, at the beginning of the year.

The significance of those numbers on the running course we don't believe is going to be significant, but of course, we need to do that evaluation every year -- every quarter.

Coming then to the cash flow element. I put up here a cash flow reconciliation table. And starting with EBITDA, cash flow from operating activity and then equity-free cash flow. This $804 million as you can see on the middle of the page with the yellow brick around it, is really going to be the basis of where we're going to start guiding from now on. The difference between the $804 million and the $1.067 billion is $263 million, which we have disclosed to you every quarter within the attachment to our report every time, which is then exceptional costs in different shapes and fashions that we have sort of adjusted just to give you the elements of where we want to go.

I think the major point in actually this evolvement in cash flow is to go from 2015 in these numbers on equity-free cash flow. You see that the number is negative with more than $800 million. In '16, it's negative with almost $300 million. And this year we're actually coming up to almost $800 million. So you see the trend line of cleaning up a lot of different elements that has been there, that has been paid through both P&L and balance sheet items during the course of the year. Because the underlying free cash flow has been quite a difference because of the major adjustment factors.

So here we are going to go with this equity free cash flow because now we actually do think that we're coming to a much normalized fashion, and have normalized both the structure and the operation going forward.

When it comes to the cash upstream developments, VEON had structural constraints by, I think, both Ukraine and Uzbekistan has sold or opened up so that we're actually not going to see too many challenges for that going forward. It is, of course, going to be restrained on how much at a certain point in time we can actually extract from many of these countries. So there will be a continuous dialogue with all the central banks in these countries. But as of now, we have seen that the improvements both in Ukraine of the cap increase over time is now -- we are not able to upstream an amount close to our annual profit.

And for Uzbek, we see that the [caps] for upstreaming capabilities will continue as we can see it now. And when it comes to the voluntary restrictions on Kazakhstan and Kyrgyzstan, that has also been removed. And then it's, of course, the GTH structure, and the MTO for GTH, which is more a technical or structural challenge on the upstreaming more than the control of upstreaming cash as such. So all in all, I do think that a lot of the challenges that has been there in cash upstreaming has been removed as such. And as a result, coming back to Jean-Yves' statement, that now many of the countries are contributing the upstreaming cash compared to a couple of years back, where it was very limited, mostly limited to Russia.

If we then go to the net debt evolution. We're coming from $7.2 billion in 2016 up to $8.7 billion. There's a few elements going into this. Of course, the EBITDA and the working capital, finance charges and taxes, CapEx and then, of course, dividends, both our own dividends to our shareholders, but also minority dividends in GTH as well as in Pakistan. The FOREX element is mostly the translation of the Uzbek devaluation that we announced in the third quarter. So most of that is coming out there. And then, we're coming up to the $8.7 billion. And as a result of the top, as a result of doing the MTO, we had to make a collateral for the guarantee of the MTO. And as a result of that, in our accounts at year-end, even though the MTO is not finished or not yet launched, we're actually going to show you a net debt of $9.7 billion at year-end.

The rebalanced capital structure. As you can see on the top here, the level of guaranteed and PJSC loan is actually coming down quite dramatically, and we have lifted now most of our debt structure up to the headquarter level. And as a result of that, we are more flexible and we can also fund ourself cheaper than we have been doing before.

If you look at the lower part, you see that the currency mix and the currency structure is now also better, combined and balanced. And as a result of that, we actually think that we're currency mix, relative to the VEON coin, is better aligned than before. We're not going to be good as a result of being in the portfolio -- as a result of having the portfolio we have, we're not going to be able to mix that currency mix in an ideal fashion due to the fact that it's almost impossible. But I do think that we're coming much closer to less risky currency mix than it was before.

Going then to the financial targets. On the total revenue and underlying free cash flow, we're delivering in excess of what we have expected. But when it comes to the underlying EBITDA margin, as a result of a challenging fourth quarter, we are not able to actually deliver on flat to single-digit accretion on the margin side.

Going then to the guidance for -- or the targets for 2018. We have here made a small bridge. And the reason for that is that during 2017, we have made certain strategic decisions and that is the Deodar transaction, the Euroset and also the $106 million we received from one of our vendors. That has a fairly significant effect year-over-year. And as a result of that, we have developed a pro forma baseline just to show that on the continuous basis year-over-year to like-for-like, these pro forma numbers is the baseline starting numbers. So to give you the best perspective of how our operations are developing, that was our best way of showing it to you. And the pro forma is then starting point is 9.350, the 3.2 in EBITDA and close to 4 -- $500 million in equity free cash flow. And then our guidance comes from there on. The guidance is flat to single-digit organic growth on revenue; flat to single-digit growth on EBITDA organically; and then cash flow and equity free cash flow of around $1 billion in 2018.

And Jean-Yves have just said it, so the board decided on the dividend of $0.28 per share for the fiscal year of 2017, and we have paid $0.11 in an interim dividend, and that leaves $0.07 (sic) [$0.17] per share for a final dividend to be paid in March. And when it comes to the statement, the policy as such, the policy remains the same. So with that, I'll finish my part of this presentation, and I'll leave it for a short break before we head on to Russia.

(Break)

Okay, welcome back from the coffee break. Since you all are younger than me or look younger than me, we can all remember that when we were students or when we're going to conferences, we had a big lunch and then we come back and we struggle to stay awake. So that's why we kept the food service very limited, so that you would be able to focus on discussing Russia with me, because I've been looking forward to that. To those who don't know me, I'm Kjell Johnsen. I'm Head of Major Markets, which means that I'm looking after Russia and Italy in this system. And I also spent the last 14 months as the CEO of Russia, and not by design, but it was a fantastic experience. I'll get back to the management changes in Russia that we have now successfully executed on.

But let me take you through some of the strategic priorities of Russia. This is a well-known illustration for the management team and the employees of Beeline, they have seen for quite some time. One of the key elements that we have focused on, at least in my time, has been the ARPU, turning around the top line, starting to drive positive revenue growth. And the key element that we have been successful is with is to drive mobile service revenue growth and change the pricing paradigm in Russia. Moving away from unlimited price plans to data-centric pricing, which is absolutely a fair approach, because you should pay more when you consume more, and it's also the right approach in terms of the spectrum efficiency and network efficiency.

Then we made a big analysis 1.5, 2 years ago, looking at the distribution in Russia. And my clear opinion on that is and was that Russia was caught in a multi-brand and alternative channel system that does not help us move into the future. Very much focused on selling SIM cards, driving volumes. They are not driving value, not driving customer experience. So we set out to change that. And that has resulted in the Euroset transaction that I will talk a bit more about later on.

Where we have had issues and where we have still not come to the final stretch is on fixed-line. We are doing a lot of positive things within the fixed-line, but lead times are big. So 2018 is the turnaround year. I'll take you through some of the initiatives that we have taken and what that gives us already now. Then there's a lot of talk about what's the right CapEx level for Russia? Should it be higher, should it be lower? Most people are wondering what's the optimum level of CapEx. And obviously, like Jean-Yves and Trond have talked about, it's not just a number in itself. It's how you spend it and it's how you structure your procurement. We are very comfortable with the CapEx level in Russia today, where we say that 15% is enough. We can run Russia at 15%. What has been much more important over the last year or so or 2 years maybe is how we structured it to work with our vendors, our logistics and these kinds of things. That has limited us more than the CapEx in pure ruble numbers.

Then, of course, to talk about digital. Digital is an area that encompasses lots of things. What I will take you through today is our thinking around the self-care, the marketplace and how we are taking the initial steps to become a leading digital operator. And I'm not going to talk only about things that's going to happen in 1, 2 or 5 years. I'm going to talk about things that we're doing as we speak.

Let me start by first things first, revenues. If your revenues are not there, you can do whatever you like, you'll be in trouble. But that was a simple starting point that I took in Q3, Q4 2016. What did we have at that time? We had a data pricing that was completely open-ended. People had unlimited tariff plans and it's a fundamentally wrong principle to build a business. It's also the wrong way to build a society, which is a bigger issue. We see now that we are moving away from having a huge proportion of migrants in our customer base to going now towards people on bundled tariff plans, which is a big shift in the market. We were also very vocally talking about the need to go to data-centric pricing plans. There was a little bit of hesitancy in the market, but as of the beginning of 2017, I think we can say that all operators have started to embrace this approach. We had a hiccup in November when MTS made some moves in 10 regions of Russia. We do believe that this is a temporary, should I say, fluke, and our response to this has been absolutely surgical. Only in those areas, time-limited reactions meant to show that we are responding, but that we are not trying to start a price war. And I think we will get out of the woods on that one. I also think that Tele2 will have to focus more on value going forward. They made a big splash of trying to drive a position in Moscow. Again, moving away from migrants, high churn over to bundled tariff plans, data-centric pricing. This is not rocket science. We have seen it in other markets and some of us have been part of that journey before. And that is the platform for future growth in mobile.

Then fixed-line. A more troublesome issue. If you go back to 2007, there was a company called Golden Telecom, very successful, growing very fast, double-digit growth every year and profitability of around 40% at EBITDA levels. This part of the business has not been given the attention that it deserves. We do see that connection times for new customers have gone all the way up to 60 days and nobody wants to wait 60 days. So the first step we said to ourselves is that we need to revamp the whole business, need to build new city rings. We need to fix our provisioning. We need to get out of having SDH networks. We need to build new IP networks everywhere. That's the only way we can drive this business. And we want to connect customers, as a first step, in 10 days, not in 60 days. Not saying that's the ultimate objective, we want to be even better. This is happening as we speak. The first new city rings were built end of last year, October and November, and it's ongoing throughout this year. So 2018 is the year of really revamping fixed. And this is important for us. Obviously, mobile is much bigger, but the decline of 10% in the fixed dragged down the whole business. We can't have that going on forever, and we won't. Where we've had success in this area is by using the capacities we have for an FMC offering. We've had a very successful FMC offering. Now with 870,000 customers connected, we want to expand that to over a million by the end of this year, and we have taken the lead position on FMC. Very successful also in terms of the go-to-market and the way we have been campaigning out there. Happy to report on that million coming up.

And then a slide that I think is super important, because this is a paradigm shift of the Russian telecom market. We had a lot of our sales going through Euroset, (inaudible), alternative channels, mom-and-pop stores, all these kinds of things. That was nice in 2005, maybe it was okay in 2010, it was wrong in 2015. And by 2018, we will finish a complete integration of not 2,000 stores that we take over, but around 1,650. We will shut down around 350 for geographical proximity. So that's just a rationalization of the market. And then we will have built the perfect distribution for 2017. And why do I say 2017? It is because we will then have the capacity that we need with the current model of distribution in telecommunication in Russia. Going down a couple of years down the road, we need to accelerate the sales through digital channels, the online channels. And I foresee that the trend when we get to 2020 and onwards, will be that we see all operators de-escalating in distribution. There has been a war on gigabytes in terms of the advertising. There has been an escalation and a war on having distribution capacity, both of them dysfunctional. This will then turn and I see MTS already starting to reduce the number of stores. As we progress, I think that the number of stores we have will, year by year, be reduced. We need 4.2 people for each store on average to keep it operating. So you can do the math. We're integrating 1,600 stores, that's close to 7,000 people. And we are very happy to do that. But down the road, we need to go to a more efficient model. And I think the whole industry is getting there, if we listen to what my colleagues in the other companies are saying. What we are doing now with integrating Euroset has probably not been done anywhere else in the world. I don't want to look at China, because China is so big in all kinds of respects, but this is a huge project. I'm very happy to say that the team is confident that most of this is done by the end of Q2. Maybe not everything, but most of it. And I have booked a table in Moscow on the 11th of July to sit down with a guy who is responsible for integrating Euroset for dinner. Reservation is in his name and he has to pay if we don't come for that dinner. So he has every incentive in the world to make that happen.

More important than the time line is the quality. We see the people working in shops now are not very motivated. Why would you be super motivated to sell a Beeline product if the only thing that drives your decision is the amount of incentive you get for doing so? We want to have people working with our products, mobile, FMC, whatever we choose to sell in that shop, and want them to be trained, and we want them to be proud of working for Beeline, staying longer with us than the attrition rate that we see in this industry today.

And then into smart investing. There are a couple of things that form the platform for smart investing. One thing is that we need to know where our biggest consumers are, our most valuable customers. And the smart way of doing so is of course to have bundles, where you see where the biggest consumption is and that people actually pay for the amount of data they use. When we have data-centric pricing and DMP in place, I will get back to DMP, data management platform, we can see these things. And we can allocate our investment resources in a good way. This is also an efficient way of building networks. We don't want to just throw out base stations and repeaters everywhere, because people like to consume data. No, we want to build networks in order to serve profitable customers of Beeline. And I think we've been pretty good, not perfect by any stretch of the imagination, but pretty good at that. It's one of the ways that we've been able to keep the CapEx levels at an acceptable level for us. In 2016, we had major issues with a couple of vendors. And I'm not going to blame the vendors for that. This always takes 2 to tango. And the reality here is that some of this could have been avoided. We had some very tough discussions. We reallocated market share. And in 2017, we doubled the number of base stations we built. That was absolutely necessary. And I'm happy to say that by and large, we have a good swap and rollout operation going on now. There are a few issues, but they are much smaller than they were 12 to 18 months ago.

For 2018, you can expect more of 2017, not 2016. We have pretty good control over the network rollout. A side effect of cleaning up these logistics that we had there is of course also that we have been able to reduce our working capital quite substantially, which is always a nice bonus to bring with us.

Then to addressable costs. We work and operate in an industry that has a lot of pressure from authorities, and you know countries, most countries, like to fill up their coffers, and there have been some tough times in many countries. In Russia, I would say by and large, the regulatory regime has been pretty good for this industry. But we do see pressure on certain taxes, spectrum fees are going up, property taxes are going up. There are always these elements that come and push our cost base. So the only way to counteract that is, of course, to work with the government to explain the consequences, but to control our addressable cost. In 2017, we saw an increase of 2%, which is still significantly below inflation, but not good enough. 2018, we need to get to flat or reducing addressable cost. So if we could end up reducing our addressable cost by 2%, it will be a good starting point.

And we will drive these programs and these initiatives throughout the organization. One effect of the changes we have done in terms of redesigning Beeline is that many people have moved over to other relationships. We are now working with Nokia and with Huawei on what we call internally the Phoenix Project. So a lot of our technical capabilities have moved out, 3,300 people got a new home through 2017, at least a new place to work, if not a new home. So meaning that a bigger part of our employees are customer-facing, are meeting employees, talking to -- sorry, to customers every day. So we say here 72% will be directly in contact with customers in their everyday life. And of course, people who work a little bit further from customers have a less high frequency on this.

One thing we have worked on quite a lot is the interconnect. The interconnect system in Russia is out of sync with the rest of the world. We still have the same interconnect rates as 10 years back in time, and paying 95 kopeks to terminate a call in this day and age makes absolutely no sense. It's much higher, almost in every way you compare to any country in Europe, and it leads to the strange situation where the wholesale cost between the operators is higher than the end-user cost. It doesn't make much sense. We are -- we have tried to address that from a regulatory point of view, but instead of waiting for that, we have also optimized our packages and our pricing to the point where we have been able to halve that leakage, primarily towards MTS.

So where are we now? I'm pretty happy with what's going on in the mobile business. We do see we have turnaround, starting to increase ARPU, driving some growth. We see a relatively flat customer base. And I would predict that the total sales of this industry in Russia will go down dramatically. It's absolutely dysfunctional to sell 115 million or 110 million SIM cards every year. So I think in stages, you'll see that coming down. That doesn't have to affect the number of mobile customers in each of the operators. It's just a way of taking away that meaningless churn that we've seen. If you look at the fourth quarter, you look at our EBITDA margin, they are down year-over-year. Adjusted for our launch in VEON and a couple of HR issues that are one-offs, it is still marginally down, but more or less on track. There is one anomaly on that EBITDA that I would point to. That is actually the month that made a difference was October. We launched VEON with a huge ATL back-to-school campaign in September. We did see some weakness going into October. We took some corrective actions and redistributed, and we picked up again and December was a pretty good month for us. So that's a granular view of this quarter. So adjusted for that, I think we are more or less on a flat development for EBITDA. And of course, going forward, as Trond has talked about, we will have the Euroset integration temporarily, of course, influencing that.

On CapEx, we keep to a very tight CapEx management, so that when you combine EBITDA and CapEx, you see the cash flow from the business is increasing. And if you add up then also the working capital improvements, there has been a fairly substantial increase of the money coming out of the Russian business. This is just a picture for the full year, where you see we get increase in revenues, driven by mobile, but of course, held back by fixed. That's why it is absolutely important that in 2018 we come back to a point where fixed is not a drag on performance, but starts to contribute. Again, mobile customers more or less flat year-over-year. And for the full year, you see that the numbers are down to the decimal point on EBITDA. So you could say the full year delivered a pretty healthy EBITDA percentage and absolute return. CapEx, coming down a bit, even though we are building much more. So it shows that the procurement argument that Jean-Yves used is a good one. It shows how important the logistics are. And again, producing more with less CapEx. It's not the money itself that was the biggest limiting factor, it was the way it was executed.

Then back to digital. And we talked a lot about our launch of VEON and we have extracted a lot of learnings from this. So one of the steps we are taking now is that we are going to develop a self-service platform based on the Beeline platform, which is a good one. It has won lots of awards. It's consistently rated very well in the app stores. And we feel that we have the best offering in the Russian market today. But that's not the standard we want to live by. We want to run My Beeline as something that is at the world-class level. And the team is now working hard on this. We will see significant improvement and upgrades of this platform during the next 6 to 9 months. It will be very tangible. What we have at the bottom here is the data management platform. And you have been listening to executives in mobile companies for many years talking about the wealth of data that we have and how we have more data than Google and we have more data than Facebook, and name all these guys who make a lot of money. But what people have been talking about is the potential. This is a reality now. We are in a situation where we are actually able to use this platform today for our own purposes and for working with partners. So there are tangible decisions -- business decisions that we make that come from the data management platform and there are partnerships. We see some priority use cases. This is obviously at an early stage. And we have been through some iterations. This is starting to get pretty stable. And these relationships are pretty solid. We have worked with X5, the biggest retailer of Russia, for a year. Very good relationship, very good cooperation, initiated by Igor and me after we had a meeting a year ago.

So what does X5 get? Well, they get a lot of location data, they get a lot of information about customers and they want to use this kind of information so that they can tailor-make their offering. So when a customer goes on his mobile phone or her mobile phone to buy a product from them, it could be something as simple as what we put here, the dinner suggestion. You use the information about where they are. So you know if that shop is close to a discount retailer or to an upscale retailer. You use the location, based on the social demographic data, to tailor-make your offer. The pricing for them can be different from store to store, based on the information they have, which is not unusual for retail stores. The thing here is that you can do it real-time. We work with GetTaxi. We have more accurate location data than which you normally would get from a GPS. There are certain areas where the deviation is pretty big. I have seen it myself many times. And we have good, developed relationships with insurance companies and others. And how do we extract -- how do we get money from this? Well, depending on what the transaction is, we can get a few percentages or we can get even 30%, 40% of a specific transaction. We help these companies in their go-to-market by taking down their advertising and their customer acquisition cost substantially. They share that with us and boost their business. This is not going to be RUB 100 billion next year, don't get me wrong. But it's already delivering tangible money to us and we are going to drive these kinds of models quite actively going forward.

And then to regulatory. I touched upon how governments in many countries are increasing the cost base for the industry. Yarovaya law is, of course, an extremely important element in the Russian business. I think the industry and Beeline has worked pretty good on this. Go back in time 2 years, and we have these apocalyptic predictions of having to store 3, 4 years' of data and even streaming. That kind of capacity for storage doesn't exist in the world. So the industry has worked and our GR people have worked effectively to take that down now to 6 months of storage for voice SMS and 1 month for video. And we're going to continue to work with the government. There are many people who have a deep understanding of these problems and who, I shouldn't say support our case, but are open to listen to our arguments. And I think we just need to push -- continue pushing that to drive down the cost of it. We want to work with authorities to ensure public safety. No doubt about that, but we don't have to do it in a very expensive way. We expect more info to be available after the presidential elections, but of course, the outline is starting to become more clear.

When it comes to national roaming, on-net roaming, I personally spoke out as the first telco executive in Russia in December 2016, that Russia is one country. And of course, when the telecom industry in Western Europe offers you free roaming when you go from one country to another, you can use your voice, data, SMS, these things, when you move between countries, it is strange that within one sovereign state you would have multiple different systems in place. There is a history to it. But I felt that, that is natural. And we are moving in that direction. We have been among the first to take those steps. Customers like that a lot. What we have asked for in addition then is that the interconnect rate is reduced to a meaningful number. That would be beneficial to us, no doubt. And we have taken out half of that cost, but we couldn't be sure about that in 2016. So we wanted to hedge our bets. But this is, anyway, the right thing to do and we will continue pushing for that. Then, spectrum. Spectrum is, of course, a scarce resource. And if it hadn't been scarce, there will be no pricing power in this industry. So it's a both good and a bad thing. Where we are now, and depending on the region you are, some regions we are absolutely happy with the spectrum situation and others we are tight. So we are maximizing the use of our spectrum. We are re-farming, actively moving spectrum bands from 2G to 4G, from 3G to 4G and doing everything we can to rationalize that. Going forward, when we look into the 5G world, we will need to free up new spectrum. The current spectrum available in Russia will not deliver an efficient 5G experience or rollout.

So of course, the 3.4, 3.6 and these kinds of spectrums need to be made available. The good news is that when we change our networks in Moscow, we all get ready for 5G, but we don't desperately need 5G in '19 or in '20. So I see this more as a '21 exercise. Time will show. But spectrum is, of course, the message to the regulatory bodies. You need to work to free up the space.

And then a few words about the management changes. Those of you who have been analyzing Beeline for many years will know that there have been too many management changes there. CEOs come and they stay for 18 to 24 months, and they move on. And that is not good for any company. So when I came in there in September, I said from day 1, September '16, I'm going to be here until I've found my successor. There's going to be an orderly transition. I'm not going to stay in this role, which is not optimum, but that was the case.

And it took a bit longer than I thought, but we have found an excellent CEO for Russia in Vasyl Latsanych, who knows the business in a very good way after many successful years at MTS, has a broad experience as a leader. And of course, we have had the change on the CFO also, Fabrizio, coming out of VEON. He was always destined to work with me with Major Markets, but stepped in as CFO of Russia in the interim. Beyond talking about Vasyl and Andrey, who -- Andrey Larkin, the new CFO, who came from the insurance industry, very, very talented and smart guy. The good news is that it's not all about them, or about me or Fabrizio. There is a very strong team in place. And people who have started getting to know working with each other, who have been through a couple of programmatic things together and now are seeing success. So what we have in place in Russia now is probably the most coherent and strong management team we have had in many years, and that gives me a lot of hope going forward.

That was basically what I was planning to say about Russia. And as I alluded to when it says Major Markets, that means Russia and Italy. So now we are kind of moving the focus towards Italy.

Having stepped out of the CEO role of Russia, I'm obviously going to spend more time with my friend Jeffrey in Italy, and I look forward to that. A little update on this situation there. We have a very well, ongoing relationship with our partner Hutch, who owns 50% of the company, both as shareholders and in the board, where we have equal representation. And we have been led now by Canning Fok as the CEO of Hutch, being Chairman of Wind Tre. In accordance with the agreements that we put in place when we set up the company, that chairmanship will go over to a VEON-nominated board member as of April. So after the April meeting, I will take over as Chairman of Wind Tre and dedicate quite a lot of time to the Italian operation. Of course, working very, very closely with Jeffrey and the team we have in Italy.

So Mr. Hedberg?

I'm actually delighted that Kjell will be our chair because many of the issues that he's just articulated we will be confronting in Italy, whether it's fixed mobile convergence, what we need to do in terms of allocating our capital in a far more disciplined fashion, what we need to do in order to revamp our distribution efficiency, leveraging digital both internally and externally, Kjell will bring a lot of that experience to bear. So it will be a good partnership.

I think what I would like to do today is really to -- after my 7 months on the job, to give you a sense of the market dynamics in 2017. I'd like to give you a sense of what we did to address those market challenges in 2017, share with you the results of 2017 and then really spend some time giving you a sense of what strategically and tactically we will be doing over the course of 2018 to address many of the challenges that are confronting the business.

And in that spirit, let me start with the market. It's been tough, and I think consistent probably with many of the presentations that you've heard from our competitors at TIM and Vodafone in Italy, it's been a very challenging year in Italy. I think a lot of that is a little inconsistent with what we learned in our microeconomics handbook, where you would think that if a market goes from 4 to 3, that there would be more of a rational behavior, particularly with the new entrant on the horizon. We didn't see that. You also would have thought that after the years 2012 to 2013, when there was a pretty substantial price war in Italy, which reflected about a 15% drop in value, that a lot of the competitors would have learned the importance of rational behavior. So 2017, we saw neither of those phenomenon, and it has been a quite challenging year in terms of pricing. Clearly, and I'll come to that in a moment, what we are trying to do is not continue to lead on price, because I think maybe, and I've lived and worked across many different geographies, when I got to Italy, it was pretty like whose market share is bigger than the others. And it wasn't about value, it was about volume. It wasn't about net adds and retention, it was about gross adds and acquisition. And so what we will be doing is giving you a sense of what that dynamic is, but also share with you some of the philosophies that we're changing in order to position ourselves for the future.

Just a couple of points on the consumer mobile space, which obviously continues to drive the preponderance of the revenue in our business and in the industry. It was very aggressive in terms of underground to tax, below-the-line offers, even the most rational, I think in the past, Vodafone, over the period of 2017, reduced its pricing and generosity by about 33%. A lot of underground activities through outbound logistics, selling services at substantial discount. And not just generosity, that was part of it, but price. And a washing machine mentality, and I think it's reflected here, 20% MNP versus 2016 increase. So 12.7 million porting requests in 2016 and it went up to over 15 million porting requests. So churn, washing machine, competition on pricing. Obviously, as we bring our networks together, our systems together, our channels together, our processes, our structures, quality wasn't great as we did that. So we were forced to compete, not only for competitive reasons, but also that the quality we had was really a differentiator. But you see again, price per gigabyte, and not just generosity, it's a fraction I know, but price was really the key campaigning tool.

Not only in the consumer mobile space, but also in the consumer fixed space, we see a huge requirement, and I'll speak to that in more detail as we move forward, to move from this focus on consumer mobile to a focus on fixed and fixed mobile convergence. But very competitive pricing there as well, as you see, 15% to 30% discounts that were being offered, as the race to move from mobile to fixed to provide fixed mobile convergence as a differentiator, was a key facet within the market in 2017. Business similarly, very price driven, not much around quality, not much around new security, cloud, number of other IoT applications, but largely price-driven in 2017. TIM had an offer called TIM Europe, which basically was providing for EUR 25, 5 gigs to its small and medium business customers. What they did is they gave discounts and incentives that brought that down by 60% to around EUR 10. Regulatory impacts, we had, of course, the EUR 30 million roaming impact, but that affected, of course, all of the players at the same time and then of course, there was a requirement to renew our licenses, again as an industry, which hit us as well.

So -- but you can't moan and groan that the market is a challenge and what are we going to do? What we tried to do again with the team over the first 7 months is really try and shift the philosophies. A shift again away from volume to value. A shift from retention is more important than acquisition, particularly given that about 23% of our customers represent over 50% of our revenues.

So what we did, of course, as part of that shift philosophically and in practice, was looking at a number of things to harmonize what we were doing, between Wind and between Tre. One of the first things that we did was really focusing on retention again, improving tied or proxy for postpaid offers. In the first half of 2016, was around 48% of the base was tied offers. And a similar period, we moved it up in 2017, the second half, to 58%.

So more focus on tied offers, retention, locking the customer in, rather than just continuing with this washing machine philosophy. The second thing we did was with this sort of, "my market share is bigger than yours," I think there was a lot of use of the handset. And so the scoring criteria for providing a handset to get that offer, to get that gross add, was pretty lax. And so we tightened that up pretty substantially over the course of the latter half of 2017, which had an impact in terms of our handset revenue, our CPE revenue fell, as you'll see, by $83 million, which affected of course the year-on-year comparison with the pro forma company.

The other thing was restoring confidence and that's effectively being far more transparent with our customers, providing a lot more simplicity, as Kjell spoke to and Jean-Yves spoke to earlier, to our customers, that they know what they're buying. And so we harmonized a number of the things that we're doing, just beyond the customer base and the definition of that customer base, which was a harmonization or a cleanup of around 600 million -- 600,000 customers, and over the course of Q2 and Q3 of 2017. But the other thing that we did, of course, was to harmonize the offers, making them far more simple and transparent, taking out some of the costs that our customers were not happy with because we're very focused on NPS now. From a network perspective, it was a big detraction as we brought the networks together. But also, it was a big detraction in terms of offer fairness, transparency, et cetera. And so we've been doing a lot of cleanup in that area, which similarly had an impact of around, netted out, EUR 60 million on our revenue. But it's the right thing to do, particularly given, and I'll speak to that in a little more detail, with Iliad coming in, they are going to be the Robin Hood, it's about transparency, it's about simplicity. And if we are sitting there with offers that no one understands or offers that people do not believe are transparent, those high-value customers will wander.

The next thing we did, of course, fixing the operations as we bring the networks together, as we bring the systems together, as we continue to merge these 2 entities. And that's easier said than done. It's tough, I've done it before. Bringing the businesses together, I think the network detraction I noted earlier was very high, 50%. I think in anticipation of the merger, there was investment, but there wasn't a lot. And of course, when you bring the networks together, when you bring the systems together, there are going to be impacts on that. So we've done a lot of work on that, looking at our NPS scores, and I think we're seeing some good improvement, which I'll speak to in a moment, both in terms of the quality of where we're swapping our network as well as making sure that we have a lot more efficiency and less manual intervention, a lot more automation in terms of bringing our systems together. You have the Tre systems, you have the Wind systems, you needed to bring them together and we've now been in a position to consolidate and manage over 60 of those systems.

Financing, Jean-Yves spoke to as well, we were able to at the end of October and close it on the 3rd of November, reduced the cost of debt that we have in the company from 5.5% to 2.7%, which represents annual savings of EUR 270 million. So tough market, but these are some of the things that we started to do in order to address some of the challenges confronting our business.

With that in mind, these are the numbers that we disclosed yesterday. You see, obviously, as a result of the tough market challenges, as some of the handset tightening on the credit that we did, reduction in CPE, some of the harmonization that we did around our offers and how we're presenting those offers to our customer, you see a revenue erosion. We've been able to pick some of that up partially, because that's a pretty substantial top line erosion, through better management of our costs, through better extraction and accelerating the extraction of our synergies. And you see that our EBITDA margin basis points improved by 200. You also see that we were able to pretty much keep our cash flow intact. One of the reasons that affected that of course, and I'll speak to that later, is that in 2017, we invested EUR 1.3 billion in integrating and modernizing the network and we'll be spending another EUR 1.5 billion this year, but getting back to what Kjell says, smart CapEx, we're going to make sure that, that's allocated in a very disciplined way.

I think the one thing that may jump out at you is the leverage ratio. Let me give a little color to that. As a result of that refinancing, there were a number of costs associated with that refinancing that represented a reduction of our cost of EUR 759 million. We also had, as I noted earlier, the renewal of the spectrum for EUR 435 million. On the more positive side, we sold 10% of the ownership that we had in Galata, which is the tower company that we created. And we sold that, got EUR 77 million for that, and on the positive side as well, we received, as a result of the agreement with the new entrant, EUR 440 million after the receivable million in terms of the spectrum that was paid by the new entrant to us. But the interest costs were needed -- we needed to incur in 2017 and that led to a clear increase in our leverage ratio. But let me be clear, as I stand, 7 months after, to you, it's going to be about free cash flow. It's going to be about delevering and it's going to be about giving some clarity, as Jean-Yves and Trond suggested, clarity on a path to dividend capacity.

So in the spirit of being externally-minded, because I think there has been a lot of internal focus within the company over the last few quarters, let's start with the market and how we see that market evolving. I noted earlier, it's been tough in 2016, 2017. You'll note here that it will continue to remain tough, particularly with the new entrant coming in. And we see continued erosion of voice. And while we see some growth on the data side, we're going to see continued erosion on the market expectations in mobile. That's why we don't want to chase price and chase gross adds down, if we are seeing sort of that is sort of what the impact will be on our revenue and why it's important to retain rather than simply to acquire. I think at the same time, we see good opportunities to grow data. 4G/LTE is only 37% penetrated. We will be complete up to 95% by the end of 2018 in providing that coverage to our customers. We also see, as I noted earlier, albeit more of a proxy with tied offers. There is going to be more of a focus on locking customers in, particularly the high-value customer, and we've been in a position to increase our tied offers as a percentage of our total customer base pretty substantially over the latter half of last year.

Our market share fell a bit, but a lot of that market share that we lost was the washing machine mentality. And we're starting to see some improvement in terms of both the margins of our customers, because we're losing the lower-margin customers.

Fixed. Obviously, very important, particularly given the underdevelopment of broadband as compared to some of the European and other peers in Italy. We see fixed as not only an opportunity for growth within the market, because of its relative underdeveloped, not in terms of just fiber penetration, but if you look here, you'll see that only 23.2% of the market has speeds of greater than 30 megabits per second. So not only is there a growth opportunity, but there is also a quality advantage that can be derived from broadband.

At the same time, while there is a growth opportunity, it's perhaps even more importantly a defensive requirement, because as people are running into providing fixed and providing fixed mobile convergence and given that 23% of our customers represent over 50% of our revenues, we need to lock in those customers. We need to provide that fixed capacity for defensive reasons as well. You see some slight increase in the fixed market expectations, and we basically retained our market share, more or less, over the course of 2017 on the fixed side.

I'll speak in more detail about fixed and some of the defensive requirements there in a moment. The new entrant. In the road show that we did in October, it was probably question #1. How do you see the impact of the new entrant? I think, as I noted earlier, they will be coming in with a lot of marketing around trust, simplicity, the Robin Hood, the new 29-year-old CEO and he is shown with a picture of Jeffrey, Amos and Aldo, is going to be the new kid on the block who will be Robin Hood and giving back to the people. So of course, it's very important and why we took a lot of the initiatives that we did, despite the revenue impact, and driving a lot of more transparency and simplicity into our offers. We expect them to launch in the first half of, as it says here, of 2018. Probably end of March, early April is the latest expectation. We also expect them to be not only aggressive in terms of their Robin Hood positioning, but also pricing and generosity. And they will largely leverage a digital approach to the market, which I think is in one way, a good thing, because it will help us educate the market, because digital, particularly from a selling, from a top-up and a market sort of acceptance perspective, is not as advanced. And so I think from one perspective, they will help from an educational standpoint to digitize some of the thinking, which I think can only be good for us.

Of course, we're not just sitting and wondering and waiting and when will it happen, but preparing ourselves accordingly. I spoke earlier about some of the things that we're doing in terms of transparency, simplicity. Of course, where we have an ability to differentiate is around the fixed, the mobile, the convergence. They will not have a fixed base. So that's clearly something that we needed to do with our high-value customers, is to leverage that capability that we will have. Obviously, bringing handsets, but with the right credit scoring criteria, the best-in-class handsets. We have a very good relationship with Apple, with Samsung. In fact, we're the best seller, I think, at the top 3 in Europe, in terms of the Apple 8 and Apple 10 sales. So we're doing a number of things there. So we have a strong partnership to leverage in that space. And the dense distribution footprint. We will have 1.5 million monobrand stores, which compares well with what not only Vodafone and TIM have at around 800 and 900, but also, clearly, will be in this world of digital, it's great. But sometimes you need to have brick-and-mortar stores where people are coming in to see the latest handset, to understand the latest offer and to ensure that they are getting the right advice. And so we're putting a lot of focus on training there as well.

So we're ready. We have a mentality that we're not going to chase down on the basis of price. We will cede market share. I think in the experience of 3, and while I know there has been a number of 10% floated out in the market of what they want to achieve, the new entrant, in terms of market share, it took 3, how long, Stefano, 10 years to get to 10%. And I think the margins were a little better. We also have an MVNO called Poste Mobile, who has a very large, it's the post office, large distribution network. They are roaming on our network as an MVNO, and they have been around for about 4 or 5 years and don't have anywhere close to 10% of the market share.

Nevertheless, we have, obviously, despite their ambitions, a remedy contract in place. It is binding and irrevocable for 5 years. It's been well negotiated, and I think acts as a pretty much a natural hedge in terms of, and in some ways, maybe better than some of the margins and commissions we were paying to some of our channels in the past. But if I look at the terms of trade that we have as a result of that contract, there will be an opportunity for Iliad to sell 8,000 of -- or to buy, sorry, acquire 8,000 of the towers that we would have as part of our decommissioning process. They do not need to select all of them, but they are available. That would give us some revenue on the basis of the ones they choose to select. Obviously, I noted earlier that they paid us money for the spectrum that we booked in 2017 of EUR 450 million, and after the receivable EUR 440 million. But I think most importantly, beyond the terms of trade and terms of spectrum and towers, we have a contract in place that will provide us with both a floor, both caps in terms of what they can put into the market in terms of volumes, both on a national as well as a regional basis. So they can't just go dump in Milano and say well, nationally, we are under that cap. There is a cap both nationally and regionally to ensure that they are not able to come and overwhelm the network that they'll be roaming on.

Obviously, we are paid for any activation that we do on our network, not only activation from us, but also any activation from our competitors. And then finally, we would obviously be compensated for any traffic that they would be generating over our network. So while we believe that there is still an opportunity for them to come in and disrupt. While we believe that they have great experience and we respect them very much in terms of what they have done in the past, we believe Italy is a little different than some of the other markets. But we -- sometimes it's good to have an external enemy to rally the troops in terms of doing a number of the things that we need to do to reinvent the company.

This next series of slides, again, as noted in some of my earlier comments, will give you some insight into the strategic and maybe at times, more tactical things that we are doing over the course of 2018. And I'll do some deep dives. You may note some of the little shaded areas. I'll do some deep dives specifically into those areas. But as you see, we have a lot to do and it's very important that we prioritize and focus on the priorities. But there is a lot to do in reinventing this business, not only in merging it and extracting the synergies and reducing costs, and leading at customer touch points rather than in market share, but also in delivering on a number of things, as this company was largely focused in the past on consumer mobile.

So just to give you a sense on the integrate, we obviously need to bring our network, our systems together, I spoke to that earlier. We need to make sure that if we're talking about fixed mobile convergence, we need to make sure that we have the right contracts, because we're not investing a lot of CapEx into the fixed-line, but we're working on that fixed mobile convergence capability through partnerships with several players, specifically, of course, OpEn Fiber, Fastweb, TIM, as key partners that we'll need in order to be able to accelerate the rollout of fixed mobile convergence, which I'll speak to in a moment. Fixed operational processes, as you have provisioning, assurance, getting a handset or a modem out to the customer, a number of these things coming from the 2 different systems. We are working on and improving. And of course, really making sure again that we are very maniacally focused on our high-value customers.

Differentiating, I spoke to earlier, fixed mobile convergence, B2B, digital and making sure that similarly, as we move to the right, we become, in the spirit of free cash flow and delevering, extremely focused on taking cost out of the business, identifying areas where there is overlap, eliminating poor-performing channels, consolidating processes, both in a sense of really ensuring a lot better speed, but a lot more efficiency. And then finally, making sure again that we're being very disciplined in how we are allocating our capital. So moving into each of these areas in a little bit of detail.

In Italy, we will have the largest and most modernized network. We will have 21,000 sites. They will be 5G-enabled sites. It will be the largest network. TIM and Voda have somewhere between 18,000 and 19,000 sites. We obviously see that as a huge source of our ability to differentiate ourselves on that quality. We're putting a lot of money into that. I think you'll see in the CapEx slides to come, we committed to EUR 6 billion over the next 5 years to invest in this new business. And so quality is going to be a key differentiator for us moving forward.

We've done 5,000 of those sites already, 95% population coverage, I noted earlier. And we're seeing, really by the end of 2018, that going to 98% and fully consolidated, the whole network down to 21 (sic) [21,000] sites by the end of 2019, 5G-enabled.

I'm pleased to say that out of those 5,000 sites, we're getting very good scores, P3 is the testing company that looks at sort of network promoter scores. And given that it represented network in the past, 50% of our detraction, we're seeing some really good results out of some of the areas that we've swapped, both on data speeds, voice quality, and so, I'm extremely pleased that the things that we're doing, ZTE is our -- is the single vendor, will be the biggest network. They're doing a good job. It's a lot of work, but we're seeing in the areas that we have swapped substantial improvements in the network quality, which I'm very pleased about. And we'll need to continue to do that, because over the course of 2018, we're going to be ramping up that modernization substantially about 4x from what we saw in -- with the 5,000 in 2017.

Similarly, on the IT side, I noted earlier, the 60-plus systems that are consolidated. (inaudible) on the finance side, a number of things, we're doing in automating HR, internal admin processes, as we bring these 2 companies together and make sure that systems are getting integrated, modernized and made far more efficient. As Kjell was noting, we are putting a lot of focus on data analytics through the DMP capabilities. We're using Google in that space. And really using that as not only a means that we can collect a lot of data, but what are we doing with that data in terms of churn prevention, understanding where our high-value customers are, what are their requirements, how are they likely to evolve and what we need to do to anticipate those requirements.

50% of our infrastructure and apps have now been consolidated and modernized.

Fixed mobile convergence, a huge opportunity. Again, as I noted earlier on growth as well as defensively. And this just gives you some of the facts here, looking at our base. And where we've put some things in the market that are doing quite well. I mean, again, it's from a low base, but we're seeing from September, we put a new offer in the market. We've recently launched one at the end of January. We're seeing some very good growth in terms of our convergent customer base on top of the 2.7 that we have. 60% of our base on the fixed side is convergent. That means that fixed customers have both fixed and mobile. That means 40% of our customers, we can go and provide mobile services to.

Similarly, and more notably, given the size and the revenue contribution from mobile, consumer mobile, I spoke to earlier, 10% of our mobile customer base is converged. 90% opportunity, that's just within our world, not even looking at the opportunity outside and some of our competitors taking away from them, as our network improves and our service quality improves at the different touch points.

What we see is that, if you can provide that fixed mobile convergence capability, you see a reduction of 50% in churn. So while we see, given the underdeveloped, as compared to some of the European peers on fiber, on broadband, while we see that in Italy, there is a pretty low, I think, it's about 20% fixed mobile converge bases compared to the 50% in Germany and in France and Spain. There's a big growth opportunity, but also a very defensive opportunity that we need to take advantage of. To cross sell both ways between our mobile and our fixed customer base, which will help us reduce churn, it is probably the -- we have many strategic priorities or tactical things that we need to get right in 2018, that is a very, very important one.

To do that, of course, you just can't put an offer in the markets, and we're putting some things in the market that we're seeing quite good traction with on the Wind side with the family, family sharing of the data that we're providing along with the mobile and the handsets and being able to share that gigabits through the family, doing a number of things on the Tre side as well more for the youth, the techies, the data-centric type of customers that we have. But you need to get the modems out. You need to get the handsets out. You need to make sure that you're providing the service that they need. So there is a huge focus now on simply fixing the basics in terms of provisioning and insurance that will be essential for us to move from volume to value and to get this right.

B2B, so while fixed has been a bit of a stepchild and in the past, I think, B2B has also been a bit of a stepchild. There's been a lot of focus both in terms of management detention, resourcing put on the consumer mobile base. Again, as compared to other markets, we see, particularly in the SOHO, there's a small office, home office, and small/medium enterprise space, good opportunities in Italy, and that's growth. I think a lot of focus has been on the high or the big companies, large enterprises. So again, whether it's fixed mobile convergence, whether it's providing better quality, whether it's working with partners like we are doing now with Google and IBM to provide more than just price and reselling of airtime to the small office, home office and SMEs, we're really focusing on this as opportunity to differentiate, and an opportunity for us to grow in a similar space.

Businesses are not only driven by price, right? So the washing machine is less there. They're going to be very powerful buyers. They're going to expect a lot, so obviously, as part of that reinvention and focus on this area, network quality, capability to deliver the -- and provision the systems and the product on time will be key elements of what we need to do to succeed in this part of our strategy for 2018 and beyond.

Digital, I think, Jean-Yves and Kjell have spoken about that in some detail. We similarly see digital as an opportunity to improve our efficiency, responsiveness, reduce our costs, make sure that we are actually leveraging digital as an enabler to drive a cultural change within the organization. So we're not just the old-school telco heads, but we're really leveraging digital as a way internally that we can do a number of things, both to improve our efficiency and speed, but also putting a lot more power into the hands of our customers to self-serve, whether that's on buying, topping up, whether it's on answering any queries, reducing the frequency of calls coming into the call center. And the good news is that we have in Italy, already quite a good, I think, similar to what Kjell noted in Russia, we have My WIND, which has 18.4 downloaded apps within the customer base. And we have, My TRE, which similarly has 17.3.

So we have 2 entities that are also very well thought of in terms of net promoter score within the company that we're really going to leverage in order to put more power into the hands of our customers. And very importantly, I think that's another -- yet another advantage of having Kjell overseeing Italy and Russia. Some of the things that he's done with his team in Russia on the marketplace, some of the things that we'll, obviously, be doing with the VEON team will be critical in terms of us developing more of a marketplace. I think that's an area that we're not as developed as what they have in Russia and elsewhere. And for us right now, the focus is on digital efficiency within the organization. Obviously, ensuring that our customers have a self-care capability with My WIND and My TRE. And we're doing some things in the marketplace, but we're certainly going to leverage a lot of the learnings coming from Russia and beyond in that space.

Efficiency is probably a word I've used too many times over the past minutes, but it's very important. As you know, we -- as part of the first initiatives with this merger, we took 2,200 employees out of the company, largely through a voluntary severance plan, that represented about 24% of the total employees. I think it went very, very well. But there's still quite a bit of opportunity within this business to eliminate overlaps, to ensure that we are driving a lot more of a streamlined, rightsized organizational structure process simplification as well. So this will be an area that we will be -- continue to be very focused on. We're in the process now of reviewing our organization, taking a number of layers out, taking a number of -- ensuring that what was in Italy a very hierarchical structure, is becoming much flatter. And I see -- within our budget as well as beyond our budget, I see continued opportunity to improve and right size and drive organizational efficiency within the business.

Now that's just one side of the V curve. On the other side, where you're taking people out, we need to retrain the people, we need to bring in more digital talent, data scientists, number of the challenges that each of the companies in this space speak to that you've heard. The good news, that I'm very, very surprised and very happy about, is there's been a lot of work done in Wind Tre, particularly in WIND on developing young talent and working with partnerships with universities, with the business schools. People within our company, the high -- high posts, we're doing a lot in that area. So I think, the other part of the V curve is, we're bringing costs down and efficiency, it means right sizing, it means people. But at the same time, I think, we have a very good base of good talents within the organization, but they're young. We need to bring in a lot more capability, a managerial capability, and I am in the process, as Kjell has already done in strengthening the leadership team. Because one of the things, I've learned over my many years is the ability to attract and develop and retain people and partnerships is simply critical to any reinvention of any business.

Last 2 slides or 3. Yes, this gives you a sense on the far left of the ambition, and how we're driving the market's commitment that we've made of EUR 700 million synergies, that we've committed to. And gives you on the ambition side, where those are coming from, from network and IT, from commercial, from sort of the admin. And on the OpEx side, I'm pleased to say that, as we break it out, we have achieved, by the end of 2017, a EUR 167 million, which corresponds to the run rate, you can read it yourselves, that we have there. Of EUR 260 million and given that we are targeting to complete everything by the end of 2019, we're more than halfway there in terms of our OpEx synergies.

I think similarly, on the CapEx side of things, we see an opportunity to take out EUR 210 million as part of that. And so if you add those 2 numbers together, and Jean-Yves spoke about synergy acceleration, synergy extraction, I'm comfortable in recommitting to the guidance that we've given on that EUR 700 million.

And if you add on top, that was not contemplated as part of the value effect of the merger between Wind and Tre, the refinancing that we've done, we come out to EUR 970 million of value that we expect and are committing. Some of it already banked. Some of it still to do, but comfortable with what we're doing in this space, but a lot more needs to be done.

And the reason is, is that, we focus on free cash flow and that's why it's very dangerous to play the pricing game with the new entrant, if you're focusing on free cash flow. We need to modernize the network, which is also an impact on free cash flow. But to delever, you need to have cash. And you need to have, obviously, a good delevering ratio, which I'll come to in a moment on what we have as a path to deliver dividends to our shareholders.

The top left gives you a sense of, I've spoken to it a couple of times already, the CapEx profile that we are contemplating over the course of next 5 years. Some of it done. The EUR 1.3 billion in 2017. And then looks to the right, gives you a sense of what the -- both the bonds as well as the banking maturity profile looks like. Again, with a note to the fact that we have been able to get to a 2.7% average interest rate over that period of time.

The bottom looks it's -- on the basis of the agreements, what we see as our targets. So while there was an expansion in terms of our net debt-to-EBITDA leverage ratio because of the reasons, I suggested, the cost that we needed to incur as part of the refinancing and others. It has expanded, but we remain committed to bring that down over that period of time. And it gives you a sense on the far-right bottom of what that shareholder distribution capacity could look like.

So final slides. Certainly, after the 7 months, it's been challenging from a top line. It's been challenging in bringing these businesses together. There have been some things that we needed to done -- do to harmonize that has had a revenue impact. But I'm hopeful, I was doubting may be a while back when I first got there. I don't yet believe, but I've moved from doubt to hope that there will be more rational behavior within the market around pricing and looking more at value, rather than volume. I think we have a clear plan in place. We understand the challenges. We've qualified. We've quantified them. We have a clear plan and set of initiatives in place to deliver on those challenges. We're going to be measuring it like mad. You've heard from the bosses here that they have a lot of expectations for Italy, but simply so do I. So I think, we're looking at bringing these businesses together. The next few months are going to be tough. We're going to, I think, continue to be very focused on costs, synergy, being more rational in terms of how we're approaching the market. But I'm intrigued and encouraged by some of the things that we're starting to see as a result of this change of philosophy.

We know, we need to do and it's simply now about doing it. And again, it's about free cash, it's about delevering and it's about giving a clear sense of capacity for us to pay something back to the shareholders. Thank you.

Thank you. Thank you, Jeffrey. So I think, just in summary, if my voice can carry us a little bit more, but I hope that you found useful today the deep dives that we've done on Russia and Italy, and on the digital side as well as giving you perspective of our 2018 guidance and the priorities that we've set for 2018. Above and beyond that, those 6 priorities. Our focus, and our story at VEON is really to continue propelling that equity free cash flow growth story. We have an ambitious target at $1 billion for 2018 in order to ensure that we continue to deliver on the progressive and sustainable dividend policy that we announced last year.

On that basis, I'd like to reinvite my colleagues up front to be able to take your questions. I'll try to answer as few questions as possible myself, fielding them out to Trond, Kjell and Jeffrey.

First question.

Nick Delfas from Redburn. It's a question for Jeff, on how pricing is going through the base in Wind Tre. So we can see the churn rates and the net adds being work out the gross adds, but how many of the rest of the customer base has been repriced in the last 6 months, let's say? So how is the retention activity working out there? And how do you anticipate that changing, if it's all, in 2018?

Yes, I think clearly one of the key things that we need to get right is network quality, making sure that we're simplifying and making our offers far more transparent, changing the perception within the market, because, I think in the past, both Wind and Tre have been largely around value or price, respectively. We're seeing good progress, as I noted in terms of number of those key customer touch points, because, again, it's more about leading at those touch points than rather leading in customer or market share. Seeing some reduction in churn, certainly over the last few weeks, leading to better performance in net adds. I think we see better value or better margin. We track sort of the value of customers moving out or margins moving out versus margins improving in our base. So I'm seeing some good improvement, not great. Good improvement, our NPS scores are coming up a bit on the network side clearly where we have modernized. So it's not -- we're not repricing. We are basically fixing some of the things that we've seen are key detractors within our customer base. So network performance, transparency of the offer, capability within the channel, capability within the call center. And we're seeing that this approach, and it doesn't just happen. I don't have a magic wand, and I can say, as of tomorrow we are going to be value, rather than volume. There's also a philosophical shift that we need to drive not only the minds of the board, in the investors and myself, but within the company. A lot of the market has been driven by gross adds, are important you know. Have a number of gross add junkies in the company that I need to also cure, but I'm starting to see, very importantly, because that sustainable improvement in some of the things that we're doing in fixing the basics or the processes and starting to see that we're not suffering substantially from our shift from volume to value in terms of revenues, we're improving on net adds. And some of the revenue impact is the market, but also some of the decisions that we've consciously taken to improve transparency and to tighten up some of the credit score.

Is there a proportion in the market of your base that's been repriced over the last 12 months other than the gross adds, obviously, coming in on a fresh price? So I'm just trying to gauge, and what percentage?.

There was some repricing that had been done in the course of -- before my time, I think in early part of 2017 and some before then. But we have no ambition now on repricing. I think we need to first get our net -- you don't want to reprice and you don't want to rebrand until you get some of these things fixed. Because I think, otherwise you're coming out with a new idea, whether it's a price or whether it's a brand, before you fixed a number of these basics.

Since you mentioned brand, actually, is your target to have a single brand in the long term?

Our target at this point in time is to reposition the brands. I think they were a little cluttered, sort of competing with one another. We see now Tre on the medium to higher income segments as being very important positions. There are techies, youth, more data-savvy device-driven types of entities, and repositioning Wind more to the lower to the medium income segments, around value, around transparency, around family. And that's how we're positioning some of the offers that we're bringing into the markets. What's really important is that we don't have 2 structures and that gets to sort of process simplification efficiency. Two structures supporting the 2 brands. And I think we want to -- before we again make a brand decision, which I think, ultimately is probably what we need to do for cultural and other reasons, we want to make sure that we're fixing the things. You never want to launch a new brand until you fix the things that are going to be relevant and tangible for a customer.

Ivan Kim from VTB Capital. I just have 2 questions. One, just high level. It looks like in many markets, you lose revenue, you respond by cutting costs, you respond by reducing CapEx, which leads to more revenue loss, so there is this loop, which I do not really see yet, that you're breaking out of it. So if you can just tell us high level, what can be done to fix this, would be very helpful? And then secondly, on the HQ costs, there is 10%, 20% reduction planned for this year, but longer term, what's the optimum amount there? And how exactly we get to that point?

I try to take the first part of the question, if I can. I think, look, in many markets, we are maintaining, if not improving our market share. We're not losing market share in most of our markets, right? The big issues have been in Algeria and in Bangladesh, right? In all the other markets, I think, our businesses have been performing well, or very well ultimately. I think when you look at the macro level in these markets, the 20% CapEx level is unsustainable. The math does not work for investor returns. These markets growing at 2% to 3% per annum, 4% per annum at a 20% CapEx level, I don't see that equation returning any cash to shareholders. So the industry has to fundamentally change. These emerging markets are no longer in most of the markets where we operate, high-growth markets, particularly if you factor in inflation, right? So the industry has got to get to levels of CapEx that we have seen in Europe or in other places, and that's why we have given ourselves a 15% CapEx level. Now in Algeria or in Bangladesh, I don't think we have been under investing in those markets. Algeria, the issue has been the structural issue and a macro issue. And I think we've solved those problems without over investing, right? In Bangladesh, we've invested where we needed to, which was ultimately around the spectrum. So we don't believe that we're at all in a model, where we are still in high-growth models, and you have got to invest to sustain that model. Those days are gone. We're in models of an industry that is at maturity, and we've got to manage these businesses for returns in cash and not try or pursue nonquality growth at all costs ultimately. But I think that's what we're doing across the portfolio. You want to add? Want to add for Russia?

It's okay. You took it. It sounds like you're completely out of voice.

On the headquarter side, I do think that it's going to be a view of continuous improvement here. I do think that, as a result of coming to the level that we're now on the 400-plus level, we have set ourselves a target for 2018. It's not going to stop there. We're going to have a continuous improvement there as well. But they haven't set a target of where the final game is going to be.

Wissam Charbel from Farallon, First one, just if you could drill down a bit on cost. So a couple of years ago, we embarked on the performance transformation plan. We deployed quite a bit of cash to get that done. We haven't seen that really come through in the last quarter, especially in Russia. Could you walk us through what exactly has been the reason why revenue has gone up, but underlying EBITDA has gone down? Regarding VEON, second question on VEON. Could you walk us through how we should think about the improvement that that will bring in terms of is it customer value, is it cash flow? You mentioned the 1%, but can you just elaborate on timing for that? And the third question is could you just walk us through top management ownership of shares in VEON?

Do you want to?

I can start a little bit on the Russian side. In try to get into a little bit. If you look, the whole year as a whole, the EBITDA is flat, basically. And that -- there are 2 major components. One of them is that we're growing in Mobile. We are also, for last 4 quarters, for the first time since 2007, growing our revenue market share in the big 3. It's never happened in the last 10 years. But we do have a drag from fixed -- these fixed contracts, some of them going back many, many years are actually quite lucrative and profitable. They're denominated in -- many of them in dollars. So we have two effects. One is that they're renegotiated. Secondly, that those who are in dollar terms become less worth when the ruble appreciates. So that's why, I said it's critically important that we are able in 2018 to come back to stabilize that business, because it makes the whole Beeline look worse than the performance in mobile would dictate. On the cost side, I didn't want to get into very much granular detail here. But there are a couple of things that come in, in 2016, sorry 2017. One of them being the launch that we did, which was a conscious choice, and which we have talked about. We also had some reclassifications that we -- on the HR side that had to do with accruals that should have been done earlier. They're not enormous amounts. And they're not recurring. So these things will not come back in 2018. That's why, I said that by and large, if you look at it, even the fourth quarter was pretty much flat, even given the specific weakness we saw in October. November was pretty good and December was good. So when you go into '18 without me getting too much into guidance here, you should not necessarily see the Q4 as a trend break. There is no reason to do that.

If you take the cost overall element then the project transformation as we have been running now for 3 years, I do think that, it's 2 elements going into this. When it comes to the total cost side for the group, the efficiency of the total nominal number has not come down year-over-year. There's a couple of reason for that. One thing was that, when we embarked on this, our view was that the government controlled tariff cost, meaning the fees, the taxes, and so forth, the growth in that the last couple of years has grown much more exponentially than we have expected. So what we have done, on setting cost on the -- what we call the addressable cost side, has really gone to cater for the additional cost on the tax side, on different types of taxes. In addition to that, the actual administration of the project itself has also been a part of this. So as a result, as I said, during the course of the way, the nominal cost has not really come down in total, but the reinvestment, meaning that the money that we spend on digital, the money that we spend on -- that cater for the increased in cost and taxes has been done to actually do that. And that's why we are pretty sort of at least over a period of time maintained mostly, when you look at country-by-country. We relatively perform on the EBITDA market share and the revenue market share reasonably stable over period of time, except for Algeria.

I think that we recognize that our efforts on our costs are not good enough, and we have to do more from that perspective overall. And hence, some of the programs, such as new systems and others are really going to kick in, in the next 2 to 3 years to really help us continuously drive down the cost structure of the business. I think just on the share ownership, all our key executives have long-term incentive plans that are linked to share price performance. There is some 200 people, executives in the company that are linked to these long-term incentive plans. So completely aligned with shareholder value creation. And I myself own close to 0.5 million shares.

Stella Cridge from Barclays. I had two questions, if that would be okay. The first of all, on the Egypt tax dispute, could you just actually run through factually the background to that dispute, the history, timeline? What some of the key points are on the both sides? And was there any particular reason that, that came back into focus towards end of last year? That will be great. And the second question is just about the net leverage. You're standing about 2.4x right now, when you put aside the cash that you have for the mandatory offer is at 2.6x. So just what are you thinking in that sense in terms of where you would like to be over the next kind of 1 to 2 years? That will be great.

When it comes to the tax claims on the GTH, it's basically long lasting cases. And they have been through the administrative court and are in the court system as such. And we are on the process on normal course of business through the court systems. The reason why the Egyptian authorities now have raised that, and basically are trying to then escalate these situations, you basically have to ask them. It is easy to come to a conclusion. That's what's going on, and that we are giving our tender offer. But then, again, there is no -- I cannot say that this is the reason or that is the reason for doing that. So when it comes to these cases, I think that they are disclosed in the disclosure in GTH, nothing new, nothing revolutionary. So nothing has been filed from our side, nothing has been requested on their side for new discoveries or new documentation. So nothing has really happened, except for just an acceleration in the Egyptian system. When it comes to the ratio, the debt level, I do think that when it comes to debt level, I fully agree that we have escalated higher than we have seen, and also, the statement that we have been saying before. Long term, we would like to come down to the 2 level on the gearing level ratio. When we look at 2018, as we see it now, we see that the cash estimate, the cash flow estimate, that we have during this year, the net debt is likely to come some down. But I do think that when it comes back to coming closer to a gearing level of 2, we probably have to look at the longer-range of time of the 18 to 24 months to actually get down there.

And if don't, may be a following-up. So now there is $1 billion of cash which is sort of separated or earmarked. And do you feel you're at the need to do any financing at the moment? Or do you feel the undrawn credit facilities you have are sufficient. And just -- because I know the maturities are just slightly under what the remaining cash balance is for 2018. So just to get a sense of what you're thinking there?

Well. The element is, we have enough liquidity and the facility is actually capable for what we have right now. But having said that, as a result of drawing down on our RCF on the $1 billion that we put in escrow for of the MTO, of course, we would look at how to deal with that when we actually see how the MTO pans out.

Herve Drouet from HSBC. Couple of questions. The first one is, how much room do you believe you have to protect your EBITDA in case of revenue decrease, ongoing revenue decrease? It looks like some of the countries are still been challenge. You're hopeful some turnaround. I was wondering, how much room from a new well you have? I mean, you talk about systems, you can put that potentially it's rolled out on all countries, you can get 1% EBITDA margins. But any other things you can put in place to protect this cash flow generation? The second question is regarding the guidance, just to ensure especially on the cash flow. So the comparable of $1 billion is on your equity free cash flow, but on pro forma, you put that to $804 million. I was wondering, what do you expect this year to come in terms of cash outflow for licenses? We know Ukraine is a potential there. And any other investing activities that can draw cash out? And if you can quantify, at least give a kind of broad figures, what you expect there? And finally, on Italy in terms of protecting those 50% revenue and 20% of your customers. I was wondering as you upsell some of the contracts to lock them in? Or are you shifting prepared into postpaid? I'm a bit more curious about a bit more detail technically, what you are doing to protect this revenue?

When it comes to the sort of the question of flexibility or sensitivity of the guidance numbers, I mean, we are disclosing all the countries, all the major countries, all the major numbers. So I would actually suggest that you have to do your sensitivity relative to actually how to find that balance. Because it's very easy. We also disclosed the portfolio exchanges that we're actually looking at this time. So I do think that when it comes to that kind of cushion, I think, it would be wrong to actually give us any guidance relative to how much cushion or how less cushions. We actually -- the mindset -- our mindset relative to give you that guidance. The guidance we're giving you is our best belief relative to that within reasonable amount of certainty, we'll deliver these kind of numbers, the way we look at the world right now. Okay? If the world change, the world change. When it comes to the licenses, I do think that the only planned or understood licenses, we have in addition to the one in Ukraine that is already happened and the one in Bangladesh, is the upcoming in Ukraine. I really don't think that we have any other envisage spectrum issues coming or auctions coming up this year.

Italy?

Yes, thank you for the question. I think the first thing that is a requirement is simply making sure that we're not giving customers a reason to churn, whether that's bad network performance, our high-value customers particularly, whether that's things that are making them call into the call center, or not enabling them to have the self-care capability to resolve the issues. We're putting the right incentives in the channel so the customers aren't just going in and then being told to go for this new pricing model, but rather there are incentives in order to be able to upsell. So I think a lot of the, call them, the basics we're working on and seeing some good improvement. Because I think in the past, there were a lot of reasons why the customers had a reason to churn, whether it was network performance, simplicity or transparency within the offer, looking at some of the incentives within the channel, which were similarly driven around gross adds, rather than retention. And then finally, not really, having the capability for customers to resolve their queries when they needed to call into the call center. So that would be sort of the basics, if you will. The second is, we have a much better understanding of that customer base now. Who they are, where they are. We're tactically approaching them, particularly in advance of the new entrants and making sure that we're really leveraging the data analytic capability, not just to sort of understand future requirements, but understand where the money is and how we're allocating resources into those particular areas so that we can retain them. A lot more focus on tied offers now. And as I noted we're moving -- and that's really upselling, I'll give you a couple specific examples in a moment. But a lot more focus on tied offers. We've substantially improved our percentage of the total bases being tied offers through handsets, through some of the things that we're doing in advance, that an upgrade is not forthcoming for the next 3 months. So how do we get them to upgrade their handset before they have the reason to start to think about churning. Who are the -- using churn or propensity to churn models coming out of the data analytics as well. As all things that I think we've had a pretty blunt ax in the past to really understand our base, because again, it's been more around acquisition, rather than retention. Specifically, some of the things that we're doing beyond just handsets and beyond sort of understanding our base and where the money is, one of the things that we're doing now is part of our fixed mobile convergence opportunity. And as I noted, that's a very defensive requirement as well, is really going into that base and going to them with a family. So we give a gigabyte package that can be shared by the family, 100 gigabyte. So you basically lock in not just the individual but you look in the family. You give them some handsets as part of that lock-in as well. And again, it's early days. We haven't really focused on that, but we're seeing some quite good traction particularly with the new offers that we've put around fix mobile convergence into the markets, seeing some good traction over the last couple of weeks, and particularly since we launched the offer in January and also what we've done in September earlier this year. We're seeing -- and again, from a low base, 75% increase in terms of the fix mobile convergence uptake. And they're not just not doing it for pricing or we're throwing handsets, because again, I think we're being more stringent there. It's really putting offers in the market by anticipating before our customer is going to leave. But I think, maybe most importantly, now is making sure that the key customer touch points and the provisioning and insurance processes are being much better managed.

Alex Vengranovich from Otkritie Capital. Two questions from my side. So the first one is a follow-up on frequency question, I guess. So we all know that 5G is coming to Russia and it'll, I think, be like one of the first markets in your portfolio to implement this. How do you looking at the spectrum you have right now in these countries? What do you think about the potential additions to the spectrum as a response to 5G rollout? As far as I understand, you don't plan any new acquisitions this year, but maybe you can expect something next year. And for sort of licenses you're interested about? And also here, do you think some 5G sharing on this these markets really works? Are you ready to share a 5G network? Maybe it's going to be more efficient for all the shareholders if you do that. And second question on the retail network in Russia. Based on the presentation, I understood that basically you have your own agenda on the retail network, first like expansion and then potential optimization. What do you think about the steps from your competitors? How reliant you are on the steps? Or you will continue to execute on your own plan, and there is no connection between the actions from MTS and MegaFon on their retail network?

Well, I'll take 5G and ...

I can take this. It's fine. When you look at the 5G for Russia, clearly there would have to be more spectrum, as I said in my presentation. In a recent meeting with the industry, Prime Minister Medvedev indicated that to be a 2019 discussion. And expressed a great deal of understanding with the fact that this spectrum need to be made available at a fairly reasonable price. I think this goes much more beyond the discussion with the Finance Minister, because at the end of the day here, we saw on 4G, Europe falling behind the U.S. and China. And Russia, the same way as Europe. And I think it will be a huge mistake for Europe, Russia to make that same -- to do the same thing one more time. The new business models that emanate out of the new capacities and the new opportunities in the IoT area, specifically will come -- will first be developed in those countries, where you implement a properly functioning 5G. So you don't want to be too far behind. Now that contradicts a little bit with what I said, this is a 2021 exercise, I don't think that is greatly late. So the government is aware of this. And I expect that, that will be a discussion in 2019, and then we can revert to your question about what it will cost, and these things are this too early to say. When it comes to retail, yes, we are going our way. We're going our way because it's the right way. But if you look at what Andrei Dubovskov has done in MTS for many years, and I would say quite correctly so, he has built up a monobrand store system that allows him to drive good performance in the market, fully inside his own system, not being too dependent on multibrand stores. And he has publicly been saying that he expects to reduce the number of stores in his system, in their system, it's not his company. They were at some point, I think, at above 6,000 monobrand stores. I believe they can run that business easily with 4,000 stores in a more rational market. And I would expect also gradually to reduce our capacity. The little unknown factor here is what will MegaFon do, because it seems like they want to go through at least an intermediate step with integration of half of Euroset (inaudible). I believe it's intermediate. And I think they will change their mind down the road. But they will see that the cost of running that distribution, when we move to digital channels also, will not be sustainable. So that's where we are. And I think it's fully possible for us, if like-for-like that MTS could run their whole business for the 3,000 stores, maybe us at 2,500 and MegaFon something similar based on where we are now with little bit of digital development. And that's a lot of money to take out.

5G in Italy?

Yes. 5G in Italy, the government has put it into the budget, but there is a different government dynamics going on here, but it is in the budget. We expect it to happen in 2018. But, again, the terms and conditions of how that license will be awarded or how the bands are going to be looking like, how the allocations will be done are not clear. But my sense is that they would like to proceed with it in 2018. We are, obviously, geared up to understand sort of what that looks like because we don't need to have too much spectrum. I think, through the merger of the companies we are able to develop quite a strong spectrum position. And we're in discussions about where do we need to have that spectrum consistent with what's going on with Russia. But to do that is a little academic until we understand the terms and condition of that spectrum allocation. We're doing 2 trials of 5G in Prato and L'Aquila, which are 2 smaller cities in Italy. I think that we're doing with OpEn Fiber and we're partnering with Qualcomm and ZTE to really understand, to inform our case what the applications could be, what is the CapEx, what's the OpEx. So I think, that will help us to inform that decision on what we need to look for. But we're also exploring other parties within the market. And to your question, I certainly would be open to looking at sharing with others, and we've already started at least on the pilots with couple of entities sharing as a way to reduce CapEx, and again, drive free cash flow. So there is an ambition. We will know who the new government is on the 3rd of March. I would imagine that ambition will be consistent. We have a team that's sort of looking at that and using some of the learnings from Prato and L'Aquila to make an informed decision. We don't need to have -- we hope it's not just two big lots that are being allocated, because we don't need to have that, and we don't want to pay for that. But we're going to need to have a little spectrum to sort of round out some of the areas that we don't have it after the merger of the two entities.

I'm sorry, I forgot to talk about the sharing part, you asked that question. I was absolutely in a go. We are doing a lot of sharing now with MTS on 4G, as you know. We're doing a bit less sharing with MegaFon, and that's little bit down to their internal thinking around that. MTS is open for it. Now the thing here is, sharing in theory is fantastic, but we also need to work with the vendors because those features that you need to share these capacities, the prices of these features used to be almost 0, when I did my first sharing some years ago. Now they cost a lot of money. So it's not as lucrative as it used to be. Going in to 5G for Russia, I think you can see that in many places -- in many areas in Russia there will only be two networks. There was -- and some areas will have three. I think it will be complete mistake to do like Rostelecom's proposals to have one network for the entire country from a security point of view, from a redundancy point of view and from a competition point of view. But I think you will see that operators will be keen to sharing the expenses, the CapEx for building out the 5G networks.

It's Vivek Khanna from Deutsche Bank. So I have three related to Italy, and then one related to VEON, if I may. So with regards to Italy, I was just wondering, how has the net promoter score changed over the last couple of quarters and how that's -- and how each of the different brands has been impacted? So that's the first question. The second question, again, in Italy is, just can you remind me again, I think, there has been a change of network provider. You moved away from Ericsson and you got somebody else. Can you just remind me as to what the timing of that is? And are there any issues with that transition, because I think I agree with you, Jeff, I think, the network improvement is going to be the most important thing to reduce market share loss? And then the final thing is just on IFRS 6 in Italy. Just wondering as to what the potential impact could be with regards to profitability going forward? What sort of impact it could have?

Thank you for the questions, Vivek. Let me -- I can speak to NPS specifically on network, where we've modernized the network, and then I'll come to the other two. But I'll go quickly, because it's an Italy show today. We're seeing some really substantial improvements on both our NPS for data and voice in the areas that we've swapped. And one of the areas we actually saw 35% increase in NPS on the data side and 20% on the voice. In Rome and Milano, let me be more specific where the volumes are, we saw improvements and we're still going through the swap. We're almost complete in Milano, and we'll set up a little bit of ways to go in Rome. We saw over 10% improvements in the net promoter score. And this isn't something that we do, calling our customers. It's P3 that goes and looks at that. So I'm very encouraged by what we're seeing there.

On the two brands, I think, it's early days in terms of some -- I think there is on the Wind -- on the Tre side. Because of some of these harmonization, let me call them that, we've seen a 13 point improvement in NPS on the Tre, albeit from a low base. They were last there. What I can also say is that NPS as long as our new Chairman of the Renco and new Chairman of the Board is going to agree, NPS will be a very important component of our STI and our LTI moving forward. But it's probably a little early to say other than the 13% and other than the network and largely within 3 it's, because of its transparency and it will, obviously, be for both brands on the NPS score. On the second question, we have chosen ZTE to do the swap of the radio access network. They've been the one driving this integration, modernization or swap integration, modernization process. They were -- that's a single vendor. They were greenfield. So when I arrived, I was little sort of single vendor, greenfield. We're putting a lot of focus on that. So I think, it's going well. This is reflected in the NPS scores. They're building up quite well the team there. My former CTO of Pakistan, I was able to lure back -- he's Italian to run that whole project now within ZTE. So I'm encouraged by what's going on there. I think there obviously, though the wrinkles, subcons, because we obviously need to have the subcons doing a lot of those things. We need to get paid on time, and those are things that we're solving, but it's moving in a very good way. There is -- presently we're reviewing options around modernizing our core as well, and the usual suspects are involved in that. But there has been no decision taken on that. What I can say and I didn't say in the presentation is that we're really seeing, and Jean-Yves and Kjell spoke to that, a lot more capability to reduce costs by aligning our procurement, and not just get, but total cost of ownership. So both in the context of the [ran] as well as what we're hoping to see coming out of the core. Modernization, we're seeing substantially better terms, which hopefully will allow us to generate, again, that free cash flow. On IFRS, I'm very fortunate, because he has had a very, very comfortable afternoon for Stefano to answer that question, our CFO.

Actually I was talking more about the bad debt potential.

Oh, the bad debt. Okay. So it's the IFRS 9. So we implemented it. We are implementing it passing from 2017 in 2018. We adjusted our way to recognize bad debt, also considering the expectation of recovery, the receivable, even if they are not still due. So at inception, in order to be much more compliant with the requirement of the IFRS. And we, let's say, take the occasion at the beginning of this year for, let's say, the first implementation of the IFRS for adjusting and reassessing all the provision already done for our old receivable that are sitting in our financial statement. And so now we are, let's say, happy and satisfied with the level of reserves that we were able to include in our financial statement. Going forward, the way in which we are going to write off receivable is based on the aging of that and the expected possibility of cashing them in.

Okay. And I guess, my final question is a little bit on VEON. At the time when this merger in Italy was announced, one of the big advantages obviously over above improving your strategic position in Italy and consolidation -- was deconsolidation of debt, and also the ability to start paying a dividend, which clearly as you've announced today, you're paying $0.28 for the full year. One of the other things that was talked about was as you've highlighted in your presentation, the ability to get dividends out of Italy, once leverageable ratios get towards a certain level. Now Jeff's got his work cut out for him, the market is a little bit more difficult than expected, and I guess, getting to that leverage level is going to probably take a little bit longer in order to pay a dividend. So I guess, from a VEON perspective, if Italy doesn't pay a dividend, what are your views with that asset going forward?

Well, the assumption is that we are going to pay dividends. So now it's just a question of delivering on the transformation journey that we have there in terms of doing the network. Remember, we have a big bump in CapEx for doing a huge swap of the entire network in the whole country on top of a consolidation we had to do before we could do that. We are also changing the entire IT stack, which is a big enterprise we will go through. And we are resetting the whole organization. So I think, that there aren't many examples of this kind of transformation in the telco industry these days. And the fact that, TIM, in that situation chose to launch an MVNO one year before the launch of the what you call it the other player. It was an unwelcome surprise for everyone, maybe even for themselves. They do have some covenants they should probably be watching, but hey, that's where we are and that's what we are going to deal with. So the ambition of this joint venture was to move from two, number three and four, who didn't have any prospect of paying dividend to making one player that over time can deleverage and start paying dividends, simple as that.

One more question from the audience.

Irina Idrissova, RBC. So just on guidance. What do you see as the biggest risks to guidance for 2018? And what are you assuming in terms of competitive environment in Russia? And, if I may ask second question on Bangladesh. You mentioned issues with S&P in the market. Can you give us more color on what are the key issues there? And what sort of outcome will you be looking for from the regulator?

No, as I said on the guidance, I don't think there is any more or less critical elements. I do think that we have taken into consideration the starting point of 2018. We know what kind of speed we are coming out of '17 with. And I think we have taken into consideration both our expectations, our ambitions relative to turnaround, both in Bangladesh and Algeria. But, of course, also that we are actually looking at not losing any market share in any of our markets. So I wouldn't really be more specific than that in sort of trying to granulate the sensitivity in the guiding.

On the outlook for Russia, I would say what you can expect is that we will stabilize fixed, because this is not something that we are talking about now. And actually, I approved that plan already in July or August. So it's in implementation. And like I said, the first 2 rings are already up and running. All the technical plans are in execution. And we are going to make it much easier to be a client of Beeline going forward on the fixed side. We're going to continue the FMC, where we have success. And I think then you can help me in answering the question yourself. What do you think MTS and Tele2 are going to do with their positions in the market. I think given the debt situation of Tele2, if they care about their balance sheet, which I think is a good idea for B2B to do, then they will be more rational. And I think for the management team that is also more interesting way to do their job, rather than just increasing their debt. When it comes to MTS, I must say that I find Dubovskov to be a very competent leader. He has shown good results over many years and reading the picture of how to run MTS going forward is not very complicated. You follow the trends. You cut your number of stores and they cut their CapEx quite a bit. So their (inaudible) costs going down. He should have an okay EBITDA, right? It's not that complicated.

I think on Bangladesh, if my voice carries sufficiently. Look, I think, we're in a market, three-player market with a very dominant #1 player in that marketplace, capturing the bulk of the net adds last year. We believe that the government needs to address this through the regulators. As you know, there are always a series of S&P type of measures that can be taken. And we have encouraged the government to look into this matter, so that this doesn't become a one plus two-player market, but it is really a three-player market overall.

So with that, and on behalf of Jean-Yves without the voice, and the rest of the management in VEON as well as Jeffrey coming from Italy, I'd like to thank you all for attending very much, and very much the activity on this Q&A session. So thank you, very much for attending and safe returns.