Turk Telekomunikasyon AS

IST:TTKOM.E

Earnings Call Transcript 2019-Q2

[Operator]

Ladies and gentlemen, thank you for standing by. I am Gerry, your Chorus Call operator. Welcome, and thank you for joining the Türk Telekom Conference Call for the Second Quarter 2019 Financial and Operational Results. [Operator Instructions] And the conference is being recorded. [Operator Instructions]

At this time, I would like to turn the conference over to Ms. Gozde Çullas, Head of Investor Relations; Dr. Paul Doany, CEO; and Mr. Kaan Aktan, CFO of Türk Telekom. Madame and gentlemen, you may now proceed.

[Sabriye Çullas, executive]

Welcome to 2019 second quarter results conference call. We are here with the management team, and today's speakers are our CEO, Dr. Paul Doany; and our CFO, Kaan Aktan.

Before we start, I kindly remind you to review the notice on our earnings presentation. Paul, please?

[Boulos Doany, executive]

Thanks, Gozde. Hello, everyone. Thank you for joining the call.

Following our strong first quarter, we are pleased to announce a strong second quarter. I would like to start with a brief update on our strategic priorities. We continue to invest in enhancing customer experience. Our customer-focused, self-service app has 29 million installed downloads. Unique telecom subscribers using the app is 12.9 million. Upsell accelerated further with digital sales revenues over this channel more than doubling compared to last year.

Tambu keyboard, the most active keyboard in Turkey with 9 million installations, with its differentiated interface, content discovery and sharing. Without leaving the keyboard, you can do daily activities from web search, food discovery, hyper local, animated artwork to sports discovery. 275 million local original content has been shared with the main app engagements 45 billion in number. We are well progressed with new features including learning by typing, adapted linguistic engine, intelligent phrase suggestions. Also we are ready with a brand campaign management tool, allowing native content marketing with Tambu platform.

In the second quarter, we increased our share in the health care venture, Doctor Turkey, to 17% to continue -- to work also on opportunities with the health vertical.

In the education vertical, we further supported our venture capital investment in MentalUP, enhancing synergies to support Sebit product sales for home and school offerings in the country and also supporting their international expansion on educational gamification.

Also in the second quarter, TT Ventures closed its third deal with an investment in the energy vertical, acquiring 11% of Eltemtek, an energy consulting and engineering firm. It has strategic shareholders, including TEIAS, General Electric, KEPEZ, MITAS and Türk Telekom. We will be working together with them on strategic sector-specific cybersecurity projects, smart lighting, part of Smart City project, power interruption detection systems as well as energy saving projects.

Apart from investments in our verticals, we are also developing differentiated digital products and working together with our subsidiaries, Argela, Innova and Sebit.

Argela's leadership in 5G and enhanced 4.5G is very key to us. It's developing soft-defined (sic) [ software-defined ] network and Network Function Virtualization based solutions, in particular slicing technology over RAN, RAN Slicing, and which enables allocation of bandwidth within existing 4.5G network. 5G would have this feature standard, as well as SEBA, which is soft-defined (sic) network-enabled broadband access.

Argela's technologies are now being proof-tested in the networks of Verizon, Telefónica and Orange. We plan to extend this success both domestically and to other international markets. We believe that this segmented technology focus will be in particular suited to price-sensitive markets, where several operations will not be able to justify a full blast 5G investment obligation and this, hence, is a unique opportunity of differentiation by Argela .

Innova is well placed to play a key role in digital transformation program in Turkey and a few other countries, in particular domestically. We expect corporates to be investing in this direction over the next 12 to 24 months, and we will be speaking more about this in the coming quarters.

Sebit is expanding its services in the education vertical in strong partnership with National Education Ministry. Now further enhanced with teacher training programs, this is a very important development. Additionally, there's a focus on private education sector driving its unique education management platform as well as its superior quality content, education quality content, in addition to tailored private tutor programs, adaptive curricula in science and math ahead of university preparation programs.

Across the TT Group, there are 2,000 ICT professionals working with a focused project structure on noncore telco market offerings as well, of course, as supporting Türk Telekom itself to extend and grow in this conventional market. I believe that this will prove to be a very important resource base over the next 5 years for the group to be a major active player outside conventional telecommunication services, i.e., into other economic activities of the country.

I will now take you through the presentation. Slide 3. Operational performance. Number of group subscribers increased by 320,000 in the second quarter, reached 47 million. Net additions in the last 12 months was 3.5 million, reaffirming that we are on the right track with our differentiative cross-sell and synergy offerings.

As you know, we executed a balanced approach between subscriber growth and ARPU growth in our mobile and broadband business. Net subscriber additions in the mobile segment was 276,000 in the second quarter. This performance was supported by improved quality and coverage benefiting from our multiband spectrum allocations, allowing optimization of incremental CapEx, in addition to our cost-effective sales channel structure, coupled with increasing utilization of our own office channel in our well-located company buildings. This has contributed to our ability to grow market share, coupled with an improved 32% EBITDA margin delivered in the first half of this year, further demonstrating our confidence on delivering our target with a minimum 30% subscriber market share profitably.

We continue to increase our subscriber base in the fixed broadband segment with 82,000 net additions in the second quarter. Deceleration in this subscriber growth compared to last year is partly attributable to

[Audio Gap]

a good ARPU performance.

In the TV business, there's a slight decline of 13,000 in Home TV subscribers in the second quarter. As you know, we decided last year not to continue with expensive exclusive sports content, which partly contributed to this loss.

Additionally, we made price increases, which had some impact. This is in line with our plan. We saw an opportunistic gap in the market 2 years ago, so we decided to aim for a #2. We've achieved that target. We aim to hold there and now boost through further services, such as Tivibu Go services, catch-up and video-on-demand.

Pay TV remains an important part of our strategy. We will be building this further in what we call our differentiated emerging wireless home segments, i.e., data coming over the wireless mobile and entertainment coming over satellite TV, which is very cost-effective. Bar the set-top box, obviously.

In the fixed voice segment, we succeeded in subscriber retention. For sustainable revenue growth, we continue our focus on cross-sell benefiting from our integrated structure, share of customers providing consent for sharing their information, within the TT group companies increased to 78% in the second quarter from 69% last year. Triple-play drives the increase in multiple ownership among the constant -- universe those customers we consented -- who consented a year ago. Our multi-play product ownership increased from 59% last year to 64% in the second quarter.

Slide 4. This shows a summary of the second quarter financial performance comparing it with the previous periods. We grew the top line by 16.3%. Excluding IFRIC, the growth was 16.4%. Consolidated EBITDA increased by 36.2% to TRY 2.7 billion with an EBITDA margin of 46.5%, up 6.8 percentage points, compared to last year same quarter, IFRS 16 implementation had a positive impact on EBITDA.

Excluding IFRS 16 impact, underlying EBITDA margin was also robust at 43.1%, 3.4 percentage points higher than the second quarter '18, with the support of strong revenue growth and efficiency measures in OpEx management.

Our first half EBITDA has already reached TRY 5.3 billion, and we increased our EBITDA guidance to TRY 10.3 billion to TRY 10.5 billion for 2019. Kaan will go over these details in his part.

We recorded TRY 435 million net income in the second quarter compared to TRY 870 million last second quarter '18, thanks to the stronger underlying operating performance and better FX environment in the second quarter of '19. CapEx was TRY 930 million in the second quarter.

As we mentioned previously, we are working on models to minimize duplicate investments by all operators. We have several advanced workable network sharing models that are well positioned to provide win-win solutions to the sector in both fixed and mobile networks.

The expected boost in data traffic and network density, for example, smaller cells with 5G, will make for fiber network being very critical. For telecom's vast ability to provide fiber connectivity to the "most economic point of the mobile operator's respective networks" via fiber, what we call Fiber to the Site, FTTS solutions, will be extremely important for cost-effective 5G deployments.

By the way here, when we say Fiber to the Site, we aren't speaking necessarily about the customer premise. We're speaking about a network element. In that regard, we continue sector discussions under the Offices of the Ministry and the BTK regulatory authority for minimization of fixed network duplication.

We have advanced network sharing models in place with successfully executed pilot projects. So these are not only ideas. So we are well positioned in both active versions and passive network-sharing models, and we are ready to deploy the best suited model to the respective population density areas considering also the level of existing duct bank availability, i.e., measuring the incremental additional works to be done on a project-by-project basis.

Slide 5, fixed broadband performance. We recorded 82,000 net additions in the second quarter, and the number of our fixed broadband subscribers reached 11.1 million. As mentioned earlier, this quarter was relatively slow in terms of subscriber additions due to the transition to the new unlimited era. However, fundamentals to execute our strategy on increasing the broadband penetration and upselling remain strong. We will continue to drive penetration via affordable entry level penetration offerings as well as geographical and regional campaigns.

Our fixed broadband ARPU increased by 11% year-on-year compared to 2% year-on-year decline in the second quarter of '18, thanks to price increases on upsell activities. As such, broadband revenues grew 21% year-on-year, which is the highest growth in the last 9 years. Around 90% of our subscribers that we gained in the second quarter preferred unlimited packages, excluding the entry package Internet Bizden.

We see strong upsell trends in the second quarter of this year within our subscribers who are in the recontracting process. More than half of our existing subscribers preferred unlimited offers during recontracting. The ARPU of consumer fixed broadband subscribers increased by around 27% year-on-year in the second quarter this year, and the growth was 33% year-on-year when we exclude the entry package Internet Bizden. We think that wider variety will also drive our broadband revenues up with further upsell opportunities going forward.

Accordingly, we recently carried out a tariff balancing in order to increase share of high-speed offerings in our portfolio. This initiative will enable us to provide best value to our customers and upsell to higher bandwidth speeds.

In fixed broadband, we plan to shift focus slightly in favor of subscriber growth as penetration growth is key for sustained growth in this segment. As such, we increased the capacity of the limited mainstream tariffs, keeping the prices unchanged to provide better value for the entering new customers.

We extended our fiber network, and we now have 289,000 kilometers of fiber in 81 cities. We now cover, in terms of home passed, 94 -- 19.4 million, sorry, homes with fiber, and the number of subscribers exceed 3.8 million. Superior network, wider coverage and higher capacity enables us to serve as the strongest fixed network wholesale operator, meeting demands of TTNET subscribers as well as all the other ISPs. Our fiber network will be more important in terms of upsell opportunities in this unlimited era.

From a speed breakdown perspective, share of our subscribers who prefer above 20 megabit per second is 37% in the second quarter, which shows that we have significant room for upselling to drive ARPU.

Moving to Slide 6. Mobile performance. Mobile network was more competitive in the second quarter compared to the last quarter. However, we continue to perform in line with our plans. We gained 276,000 subscribers with 11% year-on-year growth. We recorded 488,000 net additions in postpaid side, and our postpaid ratio increased to 59%, up 1 percentage point just in one quarter. Our last 12 months' net addition realized 2.2 million. Our focus to get 30% market share with a rational approach continues.

Blended ARPU in this segment was healthy and increased by 8.5% year-on-year to TRY 32.8.

On Slide 7, demand for data is on a natural upward trend, accelerated by LTE capacity. Data revenues increased by 28% year-on-year and now represent 60% of the mobile service revenues compared to 56% a year ago. This is supported by LTE conversion and tariff offerings, including more data plans. The share of LTE users continue to grow and reach 50% share by 9 percentage points from a year ago. 80% of our smartphone users have LTE compatible devices.

Average monthly data usage of LTE users exceeded 7 gigabytes in June.

TV performance on Slide 8. As mentioned earlier, the business case for us going forward is different than 2 years ago, as we have achieved our aspired position in the market with 24% market share in the large screen market. As such, we now aim to improve monetization of existing base, and we aim for our channels to have reasonable content costs with strong customer demand. Instead of in-house cinema channels, we are proceeding with outsourced TV channels.

Fixed voice, Slide 9. We're successfully retaining our fixed voice subscribers, thanks to our strategy, as well as our entry package Internet Bizden and third-party benefits. In the last 12 months, net additions in the fixed voice segment were 194,000. Out total access lines also remain stable quarter-on-quarter at 14.5 million, with 402,000 net additions in the last 12 months.

Fixed voice ARPU continues to be strong at TRY 22.8. Fixed voice revenue has sustained its annual growth for 4 consecutive quarters and increased by 5% year-on-year in the second quarter of this year.

In conclusion, the company is fully geared on delivering its highly focused strategy, adapting to market conditions in our main core offerings as well as keeping an opportunistic eye on emerging opportunities in various verticals. Our investments over the years have built a strong platform for our sustainable growth. Our financial results and EBITDA guidance upgrade show that we are on the right track.

Thank you. Kaan?

[Kaan Aktan, executive]

Thank you, Paul. Good afternoon, everyone.

We are now on Slide 11 with our financial performance. Yet another successful quarter with solid financial and operational results. This has also led to the decision of updating our guidance.

Consolidated revenue's up by 16-plus percent year-on-year. This growth is primarily driven by mobile and broadband businesses, which both grew at around 20%.

On fixed broadband, as you know, we see sale usage implementation at the beginning of the year. We sell these unlimited packages or unlimited sales, we now see a clear acceleration of ARPU growth, which is 11% year-on-year in the second quarter, thanks to price increases and upselling activities, especially for us on unlimited sales.

On mobile, we increased our revenues by 19%, and same number was 11% in the second quarter of last year. Net subscriber additions were 276,000 in the second quarter, and we reached 2.2 million new subscribers in the last 12 months.

On the fixed voice, revenue sustained its annual growth for 4 consecutive quarters now and the increase is 5% year-on-year.

Our EBITDA performance was strong in the second quarter. We recorded TRY 2.7 billion EBITDA with a margin of 46.5%. Even excluding the impact of changes in IFRS, year-on-year margin improvement is 3.4 percentage points. This is the result of strong revenue growth, efficiency measures undertaken in OpEx management, integration synergies, optimization in sales channel, data growth and prudent receivable management.

The contribution to EBITDA from our mobile business is also improving. Our stand-alone mobile EBITDA margin now increased to 32%, which was just 15% 2 years ago. Operating profit of mobile is positive for the past 7 quarters. Positive change in margin came from structural improvement in this business, including the increased scale, group synergies support of new frequencies increasing data consumption and optimized commercial spending. The trend is confirming our strategy of growing mobile market share in a profitable way.

In the second quarter of this year, consolidated operating expenses increased by 3.2% or 9.9% before changes in IFRS. This is well below revenue growth and inflation.

Looking at individual OpEx components. Provision for doubtful receivables declined by 45% year-on-year. Our timely decision last year for implementing more strict credit policy rules on device sales continued to deliver solid results. Our provision expense is now bottoming. We see a rather stable growth in device business and NPL ratios have approached to its natural limit. Going forward, to that extent, should follow the trends in device sales.

Commercial costs decreased by 26% year-on-year due to efficiency measures undertaken, optimized channel structure and support from digitalization. On the other hand, we have spent 18% more compared to the first quarter of this year, mainly the result of acceleration in marketing and corporate communication activities.

In the second half, we should expect a step-up in commercial activities, especially with the start of back-to-school period. Tax expense increased by 18% year-on-year mainly due to higher frequency and treasury fees, which are in line with mobile revenue growth.

Our employee costs increased by 17% year-on-year due to increase in wages for non-unionized employees during the first half. We also provisioned for wage increase of our unionized personnel as the negotiations with the union continues. So the second quarter number should be considered as close to a normalized level.

Network & Technology expenses decreased by 8% year-on-year. Excluding changes in IFRS, growth was 19% and mainly due to increase in the cost of utilities. CapEx in this quarter was TRY 930 million, which is in line with our operating plan. We recorded TRY 435 million net income in the quarter compared to TRY 870 million loss in the same quarter of last year. The turnaround was driven by strong underlying operating performance and lower financial expenses. Our net financial expense was just below TRY 1 billion compared to TRY 2.1 billion of last year. The FX losses declined sharply on the back of a more favorable FX environment and the hedges we have carried.

In the second quarter, FX and [indiscernible] expenses was TRY 427 million, which includes a derivative expense of TRY 192 million. This was mainly due [ to the time that we ] are driven by hedge accounting, and this is particularly associated with the hedging structure used exclusively for 2019 Eurobond. Since this Eurobond is released in June, derivative expense should come down significantly in the coming quarters.

Interest expense increased year-on-year mainly due to hedge transactions we carried since last year. Post hedges, total cost of financing is around 11% as of June 2019. We expect interest expense to slightly increase in the second half as we carry new hedges and reach -- reaches its peak in 2019. However, the expected decline in derivative expenses should more than compensate that change.

We now move to the next page which is Slide 12. I will take you through the debt profile. Our net debt, including lease liabilities, increased by 2% quarter-on-quarter to TRY 17.4 billion. Net debt-to-EBITDA decreased to 1.76x from 1.86x of last quarter. And again, excluding IFRS 16 impact, the ratio was 1.71. With strong cash flow, we target our leverage to come down to below 1.5x in the next 12 months or even by the end of this year. Cash and cash equivalents decreased to TRY 4 billion from TRY 7.3 billion this quarter. This is mainly as a result of close to TRY 800 million reduction of FX-denominated debt, which also includes our 2019 Eurobond.

We are now on Slide 13. I would like to give you some highlights on our hedge position. This quarter, we started to focus more on absolute short-term FX position as a key metric rather than the hedge ratio that we have been using in the previous quarters. This hedge ratio is now becoming less impactful in a trend analysis. The level of the gross debt is decreasing significantly.

In the second quarter, we executed additional participating cross currency swap transactions with euro and U.S. dollar contracts, and the total dollar equivalent value of such contract was close to USD 170 million. And this FX exposure decreased to just below $800 million from $2.4 billion of last year's, quarter increase of $130 million in net FX exposure, it was mainly due to expiring hedging contracts and use of hard currency cash, both as a result of sizable debt reduction. It's worth noting that we executed new hedge transactions in July after the quarter closing. With these transactions, the net FX exposure was reduced back to first quarter level.

As you can also see in our footnotes, the sensitivity of P&L to FX movements continued to decline. The effect of 10% depreciation of [indiscernible] TRY 664 million impact -- negative impact in our P&L, which is 14% less than the sensitivity in the first quarter, as well despite the slight increase in short FX position, there are a few reasons for that. First, we reduced our short FX position excluding the hedges by close to $400 million in this quarter, which naturally provides more protection compared to hedges since it comes with the cash that we keep in hard currency. Second, it was some varied of new hedges are lower than the hedges that expired.

With the depreciation of lira in this quarter, we are now moving to the long hedges, that have the highest protection. You may remember that our hedges get the highest protection and the [indiscernible] rate purchase to secure. Our strategy to use cash as a tool to manage total FX exposure will continue. The share of FX-based cash was 86% in the second quarter compared to 76% in the previous quarter and 69% in last year. Let me also add that with recent transactions, we are now fully hedged the debt service of 2019 and 2020, which is now giving us much more clarity in our cash flow of the next 2 years. Going forward, we target to further reduce our FX exposure.

Looking at our cash flow generation, our last 12-month unlevered free cash flow increased to TRY 5.3 billion, which was TRY 2.8 billion last year. We have been able to reduce the net financial debt, excluding IFRS 16, by almost $300 million in the last 12 months or more than $600 million in the past 2 years combined together, and simply by using cash generated from operations without any one of the proceeds.

Strong EBITDA generation, together with disciplined CapEx, will support our cash flow positively in the coming period, and we believe that this is a very strong differentiation for Türk Telekom. Our strong cash flow will support further deleveraging, including $500 million Eurobond issuance in February, we plan to refinance less than 50% of our repayment in 2019 by providing the remaining part of the debt service from our strong operational cash flows.

We're also working to secure the refinancing requirements of 2020. Similar to this currency, we plan today refinancing for less than 50% of these payments due in 2020. Although with the successful Eurobond issue, we have been approached by debt capital market investors and loan marketplace. Our initiatives will definitely be ECA facilities synergies and also TL-based financing.

We are now on Slide 14. You can see our guidance revision. On the back of the initiatives we have undertaken and our strong performance in the first half, we made an upward adjustment to our EBITDA guidance. We now expect the consolidated EBITDA to be at TRY 10.3 billion to TRY 10.5 billion range, which is up from 10 billion to 10.2 billion in our previous guidance. We maintain and confirm our initial guidance for revenue growth, which is 15% to 16% year-over-year and CapEx for the full year being at TRY 4.6 billion to TRY 4.8 billion. We will further assess our guidance especially for EBITDA once we see the trends in the third quarter and share it with the investor community if there is any upside.

Thank you very much for your patience, I'll I hand the call to Gozde for the Q&A session.

[Sabriye Çullas, executive]

Okay. So we can open the Q&A session. Thank you.

[Operator]

[Operator Instructions] The first question is from the line of Davids, JP with JPMorgan.

[John-Paul Davids, analyst]

It's JP Davids here. Congratulations on the results. Two question from my side. Firstly, just looking at your cost base, and particularly your commercial costs, you highlight the positive impact of the channel optimization structure. I was just hoping you could talk us through that in a little bit more detail. And how are you managing to cut costs, but at the same time, maintain your commercial momentum? And then a separate question. As mentioned on this call, your debt reduction to date has been done entirely through free cash flow, free-cash-flow generation, and you haven't really touched the portfolio yet or monetized anything in your portfolio. I was just thinking, as you look over the medium term, are there any assets in the portfolio where you think you could monetize at attractive levels to reduce debt at an even faster rate?

[Boulos Doany, executive]

Thank you. It's Paul here. What I'll do is I'll take the first part and maybe a bit of the second part. Let me start with a bit of the second part relating to the idea of potential disposals. As you know, one of the key things that telecom companies today do when they are doing fixed line transformation is obviously converting, for example, copper to fiber so we have actually an opportunity there, which is in a way disposing of a dying asset being copper but actually, it still has a value obviously, into fiber. So I think we've done very well in that area. We can still do a bit more. In relation to asset disposals, as you know, real estate disposals aren't really -- it's not a very good time to get high-value there, but the company is open and the board is now open to disposal of unnecessary assets. We may have more to say on that shortly. In relation to the commercial costs, actually in the first period, you may have seen that in the first part of the year, we followed a more cautious approach in relation to the broadband part of growth while we were trying to rebalance the situation going forward. So we had also budgeted a higher cost of sales over that period. In relation to optimizing our channel, what happens basically is that we are showing a very high growth in the customer base without increasing our sales footprint and making it more optimal. So as you move things within our own premises, we're basically increasing utilization of an existing facility. The quality of it is far better than a conventional store. So we have, in that sense, we are getting better volume metrics out of it. But clearly in the second half of the year, we're going to be needing to grow to the market, and that's the reason why in our guidance, we have been a bit more cautious to say that what has happened so far is sustainable in that regard.

[Kaan Aktan, executive]

Just with that -- overall this should be -- we should not be spending less than last year in absolute terms when we come to the lira, but we should definitely be spending less in terms of the commercial OpEx to revenue ratio something smaller or lower than that.

[Operator]

And our next question is from the line of Drouet, Herve with HSBC.

[Herve Drouet, analyst]

First question is back as well on the margins looking forwards. There's been strong progress of the company on term of margin improvement. You are showing and saying the competitive environment remains pretty stable. But in Q2, we've seen for example, the net adds in the mobile segment is a bit lower than what we previously thought. So it looks like there's a little bit more -- an increase of potentially the competitive acquisition of new addition in term of subscribers. Two, am I correct in saying that on the margin -- on margin progression, especially on commercial spending, you are going to increase a bit on that side. Is it what you are expecting looking forwards? And I was wondering also if you have to choose especially on the mobile segments, do you believe your conversion into postpaid, which at the moment, is working pretty well, you can continue to sustain a better mix of subscriber in the mobile segment. And my second question is about the different plan you were -- you were talking about potential network active or passive sharing in term of potential projects under the current discussion. We've seen different operators, especially national operators, you are putting some tower site business in special vehicles. I was wondering is it something potentially you can look for? And if it's part of the discussion you have with the government, and if it's a potential outcome we might see in the coming years?

[Boulos Doany, executive]

Okay. Look, let me take that second part first. And then I'll leave some room on the first part for Kaan to tell you something. The practicality of network sharing on the mobile side first is that we've increased the passive sharing materially from just over 4,000 to around 6,000 now. So there has been good progress on that and the model of passive sharing comes from a shared cost approach. The idea of establishing like a tower company was studied in depth last year across all operators. It had a complication here, I must add, relating to the various concessions. So for example you would need to use the 2G towers, which technically runs out 2023 and 2026, so there was that technicality which resulted in that idea being dropped. Though we have to be very realistic when we speak about tower sharing companies that are -- or tower companies rather, obviously they will be shared in this case, that will be listed on the stock market because you don't really get the full value of that exit. It's rather that you dispose of the towers against a firm leaseback condition, which frankly is what we -- what our approach was. If somebody would be wanting to buy our towers, such in this case like it'd be the Turkcell Tower Company, that would have been a preference by us. But that option is off the table now due to the uncertainty of rights of use of towers beyond the [ 2020 ], 2026. Not that, that won't be sold, but it could not be sold in a convenient period of time. So the only other alternative in this case is then either lease from their tower company and that didn't make much sense to be very frank with you, so therefore, we ended up in a sharing model, which is a cost sharing model, which is the most viable alternative because you can do it with 1 operator or you can do it with more than 1. Now the active sharing, the active sharing is much more interesting because there the saving is very, very high. But of course, this would work in lower population densities. As you get into the bigger cities, we considered with Turkcell an active leasing model because obviously, we don't benefit mutually by sharing with them because we don't have much to give them in return for what we get from them. So we would offer a pay model. So therefore, the pay model would be a more suitable model but frankly, even that turned out to be also quite complicated. So therefore, we opted for an active sharing model, which we've executed now in Zonguldak. So we have 2 working sharing models. We have a passive sharing model, which is in place, which works on the principle of 1 for 1, nonpayment. And we have an active sharing model where 2 operators enter into a full active and we can convert this, by the way, into a pay model as well. So we have these 2 advanced, let me say, working as opposed to just having ideas on the table. These 2 have been executed. We also now have a, let's call it a fixed network sharing model, which is an area based, the project in Ankara that we have, which is a specific location. This works very well where the additional ducts that will be needed by the applying party is approximately something that we have something like half of it almost existing and the rest of it would be new built. This pricing model works very well for that situation. We may be able to tweak this model a bit but at least we have now 1 working model, which Turkcell and Vodafone can use on the other side. The signing party was Vodafone. So it would appear to me that the most viable approach for this particular market would be what I just indicated to you, and I am very optimistic now with the combination of this, that we can even cover it up only -- cover only a facility-based leasing, meaning to a facility, a network facility. I think also we can provide fiber to the premises of a major, for example, customers say of Turkcell or Vodafone that requires fiber to that facility for them to provide some heavy coverage internally. Okay. That would be like a corporate model, maybe on a 5-year contract, perhaps even up to 10. So I think that over this period we've progressed so much that I now see that the market will be very, very CapEx optimized going forward. And the strength of each operator will be such that we are still able to compete with each other, but we don't -- we minimize duplication. Again, this is very important to the customers to get obviously competitive pricing. I hope I gave this exhaustive answers that covers any future questions that others may have on the matter of sharing.

[Herve Drouet, analyst]

Maybe just a follow-up on this. I mean, if you will have better visibility after 2026 and let's say, at 1 point of time, there will be more visibility. Can it change the current set-up on network sharing between the operators in your view?

[Boulos Doany, executive]

No, no. Thank you for asking that so I can make it absolutely clear. The reason I refer to 2023 and 2026 on the 2G because all the others are 2029, and I'm sure with 5G, and I'm sure there will be a question on that coming, but I am sure that the operating licenses for mobile for justifying 5G will have to be 15 year forward plus. Okay. So let's say that this is done next year or 2 or 3 years forward, you have to add 15 years from that date. Okay. Now if there is going to be a tower solution, the tower solution that I mentioned to you assumed all of this is resolved, okay? And all of us will put our towers together into a tower company and that company will be listed. So we considered that scenario. But we could not -- we couldn't make it work, not due to regulations. We couldn't make it work financially between us because frankly, from our perspective, I prefer to take the cash and get a fair leaseback because I don't believe in a listed tower company structure because we don't recover the money. If we're able to sell all our towers, 3 of us, so the tower company on fair leaseback we would do a full tender, the paying party will be paying a premium to us and we would use that money to cover our CapEx on fair leaseback, that would be a better model for me than establishing a tower company and listing it on the stock market. You see? And that is what we offered but that will be too complicated to do without sorting out the ownership rights and also the rights of use, you see. So therefore if you tell me now, if you could do that, would you do it? Well, if I had the certainty of do it and the other operators would agree with it, I think we would -- all of us would agree to it. But given the fact that it's not an option on the table, I think what we have now is the second best thing and it gives us more flexibility, by the way, which is probably in the longer term, a better option. So what we are now doing is probably best for the future.

[Herve Drouet, analyst]

And on the margin?

[Kaan Aktan, executive]

Yes, okay. For the margin, I think first part of the question on margin was on mobile market. Actually, the outcome for mobile business was -- for this quarter is pretty much what you are expecting to have. We were expecting to have a certain size of subscriber gains, maybe not in the size of the last 2 quarters which was, to be honest, very exceptional quarters. But in the long-term, it will take us to a certain market share gain for the full year. And we are also expecting to see an acceleration in the ARPU growth and all of those 2 factors are justified and happened in the second quarter. The market was very strong with postpaid. As you said, we saw subscriber interest, not only for the new starters when they come to telecom but also our prepaid subscribers are converting to postpaid tariffs and we are supporting this process in the company. The margin, overall, there are a few factors. One, as you said, we should have the highest spending in the commercial area in the second half of the year. And also you should consider the normalization in a way, in the headcount-related costs. Every quarter there are more adjustments made due to the timing of the contract, employee contracts, there are adjustments, inflationary adjustments in the base, which will be fully impacting our second half headcount-related cost item. And as you know, there is also a calendarization in the commercial activity, it's pretty much busier -- busy at the -- towards the end of the year, the last 4, 5, 6 months. We wanted to see the commercial trends and also see the full impact from a few OpEx items. And as I mentioned in my notes, the third quarter result may be a good opportunity to make a fair and more sound judgment on the full year expectations for EBITDA.

[Operator]

And our next question is from the line of Ozkan, Atinc with Wood & Co.

[Atinc Ozkan, analyst]

This is Atinc from Wood & Company. Two questions, please. The first one is regarding [indiscernible] on the rate hike negotiations with the labor union. I know less than half of your employees are unionized, but I think there has been some delay there. Can you provide some color, what's the underlying assumption in the provisions for the wage hike? And when do you expect this to be finalized? I know -- I think there is a deadlock with some other unions' collective agreements with the government. Is it impacting you as well? And my second question is, I know that Paul has touched on the pay TV issue and I know you're cautious approach on expensive sports content, but there are some news that the talks between Digiturk and Turkish football collapse -- have collapsed. Will you have a look if these football broadcasting rights come to the market?

[Boulos Doany, executive]

Yes. Okay, if you like, I can take the first. Let's start with the football subject. I don't see any reason why what happens there is of any relevance to us actually in that regard. We didn't participate anyway in that particular tender. So I don't see any relevance toward that. I'm sure the parties will be reaching some reasonable solution, and I can't comment on what would be right in that regard. And I have to say also on the labor union question, we don't see any problem in relation to this subject. I'm very, very confident that we are going to be able to reach the right solution for this negotiation as this company has repeatedly been successful in so doing. These are people who are working in this company and they are also an important resource here. So therefore, I don't see any difficulty with that.

[Atinc Ozkan, analyst]

And the wage hike assumption in the provision, sir?

[Boulos Doany, executive]

Well, I mean, it's an ongoing process, ongoing negotiations with the labor union. As you can imagine, we cannot give you much details but I can say that we are properly reserved for the salary adjustments so far in place under the union.

[Operator]

The next question is from the line of Cabejšek, Ondrej with UBS.

[Ondrej Cabejšek, analyst]

Two questions from me, please. Firstly, on the mobile business. So as you're approaching your target market share, sort of around 30%, I assume, how do you think about your pricing gap or discount gap to be market leader going forward? Is this something that you're going to be narrowing down? And maybe extracting growth more from ARPU than subscriber growth going forward? And second question, now that you've got a couple of months of data from the first expiring Bizden contracts, can you just talk us through how the upselling is going and what the conversion rates are? What sort of a packages the former Bizden customers are taking, et cetera?

[Boulos Doany, executive]

So your second question again, buddy, can you just repeat that?

[Ondrej Cabejšek, analyst]

So the second question is around the Bizden product and the migration from Bizden to regular products. If I'm not mistaken...

[Boulos Doany, executive]

Okay, sure. Okay. Look, let's start with the second one. As you know very well, the purpose of the entry package was to find a way of increasing Internet penetration knowing that we may end up with a higher market share from it, but deferring any value it brings to us directly as a wholesaler and trying to grow the market without any net gain to us but also without any, let me say, clear net incremental cost as well. I think that model has worked quite well. We were hoping that competition would pick up more with these customers frankly. And I'm very hopeful now that with these guys upselling and so on, we have now around 1 million of these and already just under half have upsold. So this is very, very key to growth of the market. And this effort will be continued. We had done it on a 20 gigabyte provision, now we are increasing that to offer options of going to 40 and 60. These are now being evaluated by the regulator. So in that sense, this strategy has been -- has proven itself to be workable and the regulator has approved it after some push and shove obviously because there was a concern there in relation to this being nonuniversal pricing. It means it's based upon pure incremental at the point of having an existing home connection, meaning, we don't have any incremental cost. So we will be sustaining that in addition to having special discounts on what we call [ reasonable ] positive discrimination. So the difference now as usage rates are increasing -- when we launched this, the average was around 70, 80 gigabytes, now the average is over 100 in the market and now unlimited has become kind of like hot. But unlimited is at low speeds, you see. This is also very important. So now our target is actually to try to optimize the way forward by having a monetization model that allows the lower end to have low pricing, which is very key to maintain the penetration, otherwise they will simply be mobile, and that will be missing an opportunity if we are able to capture them at the lower cost on fixed because they don't need that much bandwidth. But also now even more importantly, is upsell opportunities with higher bandwidths, meaning obviously unlimited, also at affordable pricing that allows us to grow. Now this strategy is something that we will be working on with the regulatory authority, which will meet both ends. This requires a little rebalancing, but I think that we will be able to pull this one off because the competition benefit a lot from this approach as well, okay. So that's in relation to the fixed side.

In relation to the mobile. So when we set a target of 30% market share, this is an aspirational target in the sense that a company that has too heavy fixed and market share below this magical number of seeing a 3 in front of it, is critical as we get into the future of this next-generation of 5G. This is the reason why we need to be getting over the 30%, okay. And I think that what is so key for this company is to do it profitably, and that's the reason why the company has been very smart and opportunistic in the way it grows because it's very easy for us to grow our market share to 33% and lose money, and that would be a catastrophe, so that would not be achieving the target. So profitable growth requires us to be responsible to know what -- where to hit in the market, where we see a gap so we are very, very opportunistic. That's why in past year, you may have seen in some quarters we actually slowed down a bit, okay. So this approach has worked very well for us. And other than outlaying to you the principles of how we operate, I don't think there's anything else I can say.

[Operator]

And our next question is from the line of Ahmed, Rishad with GoldenTree Asset Management.

[S. Rishad Ahmed, analyst]

Congratulations on another set of great results. Just on the -- I know there have been a few questions on the mobile EBITDA margin. But just taking a step back, there's obviously been massive inflection in the mobile EBITDA margin over the last, call it, 6 quarters from 12% EBITDA margins in 2017 to 32% in the latest quarter. I know that there are going to be some -- a couple of headwinds in the form of higher commercial costs and higher payroll costs. But going forward, what do you think a realistic ceiling is for EBITDA margins in this segment? Because if you really look at it, over the last 6 quarters, in terms of the amount of revenue you've generated, about 95% of that has flown down to the EBITDA line for -- just in the mobile segment. So to me, that says that there is a certain amount that, that -- of infrastructure that had to be kind of laid out and now you are reaping the benefits of scale on that. But how much further is there to run on that and maybe you could just talk about the ability to raise ARPUs even further or tariffs even further within mobile currently and if that's possible, I know that was up by about 7% in this current quarter?

[Boulos Doany, executive]

Thank you. Okay, look. Again, to try to explain a principle that you can then obviously model because principles are really all I can share at this level. Now customers who are using their mobile devices for data usage are very, very interested in the gigabytes that they consume, correct? I mean, all of us are giving our customers more voice entitled minutes than they are using and definitely more messaging than they are using for those who are still using some messaging, correct? This applies to the whole sector everywhere. So the only monetization left is coming from the mobile data part and it is important that we maintain some level of competitive pricing at that cost per gigabyte that a mobile customer is using. Now the complication with costing a mobile data usage relates mostly to the coverage dilemma, correct? So you have the issue of having to provide coverage and that is where we have some competition on coverage and that's why sharing is very important. So when you get to the lower population densities, the sharing is very, very key. It can be passive, if each operator finds it more convenient to use their actual base station, or it can be active, active sharing, in which you're sharing the full cost of what's there and each one is using their own spectrum on it and that is extremely efficient if you cost it on a per gigabyte of usage. Now notice when I'm referring to gigabyte, I'm not referring to bandwidth first. The reason I'm not referring to bandwidth is because most customers on mobile are much more interested in having a low price of their consumption and they still want to consume more. The alternative of that consumption is to hand off some of it to a WiFi at a lower price, which means a lower cost. The only way you can provide gigabyte at the lower cost would be in certain locations what we can call a hotspot, correct. It means it's a place where maybe you live or you work or a third location that you go to a lot. So those locations can be covered by a fixed network much more cost-effectively. We can drive very high throughput on a fixed network at a very, very low cost per gigabyte with meaningful bandwidth almost to any location where people are living so long as you're not too far away from a city, meaning, I'm not talking of obviously rural, correct. So therefore, the customer behavior has to follow their ability to pay and our cost. Much lower cost on fixed, much -- a little bit higher cost on mobile, if bandwidth is not key. Now where are we in the market now? When we say that those who are using LTE are in the 6 gigabyte, so what that number will reach. Even if you double that number, you can still do that the 4.5G network that we have. So as long as the customer is happier with that than needing more bandwidth. If the customers need more bandwidth, there are 2 ways to do it. Either you get more spectrum and you build a 5G network which is what other operators are doing, which I think is a very high cost, which may not necessarily be able to monetize it. So therefore, what we are now getting towards here is to have gigabyte usage without, let me say, meaningful bandwidth. Let's say, around 10 megabits per second, in that range, to be very cost competitive because the cost is also competitive, to make bandwidth a premium, so that here we can do slicing, for example, over an existing network, or do spotified G. So while you are in the place that you work and you need a lot of bandwidth onto your mobile device, you cannot rely on WiFi. That is the future by the way, which we see happening in this particular market and all operators will be competing in that space. Again, there, we can sell that bandwidth at a premium but not anywhere you are. So where you are in the spot of coverage, that spot or a cluster, you'll be able to get that band because you're paying for it. And our cost of providing it is not that high, okay, because it's an incremental cost. Now if you move from that location to another, so you -- now you're driving on the road and you're a doctor and you want from your phone to still have better bandwidth than the average person, okay. There we can provide what we call an enhanced mobile service. What does it mean? That this customer is paying a premium for the bandwidth and therefore needs to be given a superior access to bandwidth. This way we can monetize our mobile network more cost-effectively, at the spot of coverage, we can give the customer anything they want because they're paying for it. And let's face it, the cost of us doing it at that location is a lot lower than saying, I'm going to give you 100 megawatts per second wherever you go. Okay? Because nobody is going to pay -- I mean, how many people in this country would pay a premium for 100 megabits per second everywhere they go. That number is not a high number and nobody is going to provide coverage for that, you see? And then, of course, the last leg, which is the hardest and the most expensive is latency. As the customers who need very low latency would pay the highest premium. So if you want to imagine in your model, capacity gigabytes, bandwidth is like in the middle where the usage is obviously very high, but they want the bandwidth, at a premium, and then the next level would be latency, and that's how we see the market, and that's how we're gearing up for enhanced 4.5G, what I call, and what will in the future be a 5G service. I hope that helps you.

[S. Rishad Ahmed, analyst]

But in terms of -- then in terms of the ability to push through further tariff increases given the current competitive dynamic, what do you see there?

[Boulos Doany, executive]

Well, obviously price increases in a high inflation market, in all businesses in this country where inflation is faced, companies increase their prices. Obviously, subject to competitive pressure, right? So in some cases, a provider of a service sees an opportunity to grow when others are increasing prices, it happens in some sectors as you can well imagine, someone who is trying to build market share. That's the reason I was speaking about discipline, you see. Now clearly, we are trying to increase our market -- mobile market share opportunistically. So we play that game where we want to be competitive, yes. So we're growing. At the same time, we're increasing prices, okay. But in the end, we increase prices where we are still able to compete, but the gigabyte is the clue to you, you see. The gigabyte, our cost of the next -- so if you're consuming, for example 5 gigabyte and you want to go to 10, right, the price shouldn't double because my cost is not doubling. Is that clear? Right? But however, however, if they want bandwidth, that customer that maybe is using 20 gigabytes, I would like that customer to pay a premium for using the bandwidth. You see, Rishad. This is the point I was trying to make to you. So then when you are monetizing your network, you can provide less bandwidth to the person that doesn't need that premium, in which case, I can pass on price increases and more gigabyte and still be competitive, while I'm using my resource without having to build another base station. I'm using an existing facility and using it to its limit. We have a lot of spare capacity, by the way, in our network, in order to accommodate future growth without incremental CapEx. But the key is bandwidth. Bandwidth is the most expensive thing to give. If you give a lot of bandwidth to everybody, you need to build more base stations, your costs go up. You see? That's the clue.

[Operator]

[Operator Instructions] Ladies and gentlemen, there are no further questions at this time. I will now turn the conference over to Türk Telekom management for any closing comments. Thank you.

[Boulos Doany, executive]

Thank you.

[Sabriye Çullas, executive]

Thank you for joining our call. And if you have any follow-up questions, please don't hesitate to contact us or the Investor Relations team. Thank you.

[Operator]

Ladies and gentlemen, the conference has now concluded, and you may disconnect your telephone. Thank you for calling and have a pleasant evening.