StarHub Ltd (CC3) Earnings Call Transcript & Summary

Hi. Good evening, ladies and gentlemen. Thank you for joining us for StarHub's FY 2019 Results Call and Webcast. My name is Amelia, and I'll be the host for this evening. We'd like to thank you also for accommodating the change in date and time of our call and webcast due to alternative arrangements being made in view of the ongoing COVID-19 situation. With me this evening, we have our Chief Executive, Peter K; our CFO, Dennis Chia; Head of Consumer, Johan Buse; and our newly appointed Head of Enterprise, Charlie Chan, who joined us in January 2020. Peter will be bringing us through a quick presentation before we open the floor to questions. [Operator Instructions] Peter, would you like to start the ball rolling?

Thank you, Amelia, and a very good afternoon, ladies and gentlemen, and thank you for the interest you've shown in our company results, and apologies for the few minutes delay in starting today's session. I will go through a number of headline messages from the 21 pages presentation pack we just sent to you. Then I'll invite our CFO to make a few comments, and we'll open up the session for Q&A as quickly as possible, which is really the objective, the key value for this afternoon session. Let me start by, first of all, saying that we have -- we've executed the strategy this year, which was based on 4 pillars, and we've got that on Slide 3, predominantly driving improvements in customer experience, and we did this through the Hello Change program, through digital innovation and through migrating customers from cable to fiber. The way we're measuring our improvements in customer experiences through the Net Promoter Scores, and we've seen right across the business an improvement across every product line in every channel. The second part of our strategy was all about trying to get more value from the existing core business. And again, we drove a number of operational efficiencies. We optimized the workforce and we launched some new services like Go Max TV product, OTT, and we also renegotiated a lot of our contracts to realize procurement savings. The outcome of that strategy is that we had a cost improvement program of $210 million. And at the end of December 2019, we delivered at least 64% of the savings. Some will be reinvested and others will go to the bottom line. Also, for some of our new products like the Go Max OTT TV service, we're reporting a healthy growth in that part of the OTT, part of the Pay TV business. The third part of our strategy was all about trying to realize growth from new investments we're making. We continue to invest in our cybersecurity business and realizing significant growth. We also announced in quarter 4, and most recently, that we're putting in a joint bid to build a 5G network together with one of our competing companies in the marketplace, and we continue to explore M&A opportunities. The outcome of that execution of the strategy is evident in terms of growth year-on-year of cybersecurity revenues, that 79% year-on-year, and you're seeing in the marketplace a number of 5G trials that we're conducting with various other commercial customers and academic institutions. And of course, one of the things we're proud of was the very first 5G COW, as we call it, Cells-on-Wheels. Finally, it is about digitizing as quickly as we can. And again, we invested in the last 12 months in digital capability by the MyStarHub application; digital sales channels; giga!, which is a totally digital brand launched in the second quarter, continues to have traction in the marketplace; a number of enterprise products and services. So this is a fundamental 4 -- strategy based on 4 pillars. And if I now refer you to the guidance we gave in the marketplace, we did say to you at the end of quarter 3 that we will try and deliver revenues between 2% to 3%, service revenues that is. We actually delivered a greater decline than that, 3.7%. But overall, our total revenues declined by 1.3% for the full year. The second guidance we had given you was the EBITDA margin on services, that after the IFRS 16 adoption, we will be delivering 30% to 32% margin, EBITDA margin. We actually delivered 31.7%. The third guidance we offered you was the CapEx-to-revenue ratio of between 8% to 9% for the year, and we delivered a 7.5% CapEx-to-revenue ratio. Some other key indicators and our CFO later on will elaborate a little bit more. The prudent management of the business resulted in increased cash flow by over 50% year-on-year at the end of December, whilst at the same time, we kept the net debt-to-EBITDA, the leverage of the company, at the same level of 1.5x. If you look at Slide 4 of the pack, let me give you a little bit of a breakdown product-by-product of what is going on in our business. The mobility side of the business is stable quarter-on-quarter. Although for the full year, this is quarter 3 to quarter 4 2019, we're having stability and a very small growth in terms of total revenues. But for the entire year, the Mobile revenues post and prepaid declined by 7.2%, and ARPUs declined by 4% to $40 ARPUs. On the right-hand side, the drop in prepaid customers was 1.3% year-on-year, but we grew the postpaid customer base and the overall Mobile customer base by 3.5% for postpaid. If you look at the TV side of the business, continues to be challenged, and we migrated the entire customer base in quarters 3 and progressively quarter 4. So our revenues, unfortunately, year-on-year declined by just over 20% to $248 million revenue for the entire year. ARPU has also declined to $44, which is about a 12% decline. And the number of subscribers at the end of December were 329,000, which also shows a 20% decline. Again, what's important in this line of business, and we're analyzing the customers migrating, a lot of customers are going to alternative, not linear TV alternative sources, and piracy continues to impact the business. The good news from Pay TV is after the migration, quarter 4 2019 revenues compared to quarter 3 2019 revenues are stable, and actually a very small increase in ARPU as well. If I now refer very quickly to Broadband. Broadband revenues year-on-year showed a slight decline of 5% -- just over 5% and ARPU of $29. Although year-on-year, the total number of customers grew by 3.94% to 501,000 customers. The drop in revenue and ARPU is part of the TV migration. We had a lot of aggressive promotional offers to maintain our bundle customers for Broadband and TV, and the ARPU as a result of the aggressive promotions deteriorated. If we look at the enterprise side of the business, our enterprise revenues grew. Our core revenues were -- grew in terms of overall -- sorry, overall, is a bit flat growth, I'm sorry, year-on-year. But quarter-on-quarter, which is a bit seasonal because in quarter 4, we do finish a lot of enterprise projects, so quarter 3 to quarter 4 in 2019, we had a 3.9% growth in revenue. But overall, the growth in the enterprise networking business was flat at $429 million. Cybersecurity continues to grow both year-on-year as well as quarter-on-quarter, and we see that growth continuing. If I now refer you in terms of overall activity in the marketplace, we still see continued competitive activity in Mobile, in Broadband as well as in the enterprise level. We don't believe in the short term that activity will be reduced. But we also see some opportunities in those markets as well. We see the forthcoming opportunities potentially with the introduction of 5G services early next year, in the preparation leading up to it. We also see opportunities in the enterprise market through a number of ICT solutions, including cloud services, data analytics, AI, and we also see the opportunity to grow our SME customers. In terms of -- if I jump to a slide which talks about guidance. Again, our strategy still is very focused in continuing to transform the business to create sustainable growth and better margin from the core business. We continue trying to diversify our revenue streams, both across different products and also geographical reach. Obviously, looking forward, we are trying to leverage forthcoming opportunities with potentially, should we be successful, to have a 5G network and also very prudent capital management. These are the pillars that underpin our growth strategy going forward. And in terms of guidance to the market, on Slide 20, and what we're suggesting to you and we take into account the current potential impact of the coronavirus because we are seeing some impact in terms of retail activity in terms of IDD revenue, outbound roaming, inbound roaming, traffic and the likes. So we believe that whilst we will continue to grow the business, the service revenues will grow year-on-year by 1% to 3%. We believe our service EBITDA margins will be reduced in 2020 in the range of 27% to 29%. And whilst we're growing in revenue, of course, the result in EBITDA is lower than the previous year because the mix of revenue, we're growing in revenue with lower margins, especially in the enterprise solution business versus mobility, so that is bringing the EBITDA margins lower year-on-year. Also, in terms of CapEx commitments going forward, we will continue to be prudent about how we're allocating capital, especially in 4G and other parts of the business vis-à-vis the potential investment in new technologies, wireless technologies. And we believe we will keep our CapEx commitment to about 6% to 7% of revenue going forward. And as a result of that and consistent with the policy that we announced last year, and certainly our CFO will elaborate, we're keeping our dividends to $0.09 at this point in time. And of course, the news are payable twice a year rather than every quarter, and that is $0.09 for the full year, not every quarter. Having said that, I will hand over to our CFO to provide a little bit more granularity. And again, we're not covering every slide. We want to open the session to Q&A. Dennis, over to you, please.

Thanks, Peter. I just want to -- and thanks for everyone for joining the call. Just want to do a few call-outs in terms of our key highlights. Peter has actually elaborated on our revenues, and I just want to do a few call-outs in terms on the cost side. In terms of costs, we have actually seen savings across the board on content cost as well as traffic costs. We've done -- we've seen savings and realized savings in staff cost, in repair and maintenance, in operating leases as well as marketing and promotion through a lot of prudent cost management activities, focusing on prudent cost incurrence, but taking into account the long-term strategy of the company as well. A few key highlights of our financial results are that we've actually reported EBITDA margins of 31.7% -- for our service EBITDA margin for the full year of 31.7% and net profit after tax attributable to shareholders of $186 million translating to $0.107 on an EPS basis as well as a full free cash flow of $218 million or $0.126 on a full year basis. Those are the key highlights of our financial results. I will pause here and hand the floor back to Amelia for Q&A.

Thank you, Dennis. We now open the floor to questions.

Let's take the first question from the conference call. I see that we have a question from [ Sachin ].

Thanks for the opportunity and congratulations on a good set of numbers. A couple of questions. After a long time in this quarter, we saw stabilization and in fact, some partial uplift of ARPU in the Mobile side. And in fact, the revenue was quite stable quarter-on-quarter, too. So what is the underlying -- I mean, do you think this has bottomed out? Or Mobile revenues are still something on a downward trajectory? That's question number one. Question number two, your cybersecurity is now $150 million kind of revenue and still loss making, does it need to be a $500 million business to be profitable or no? Is it a question of specific projects? And would StarHub also benefit from a recent announcement by government on the cybersecurity budget of almost $12 billion for next year? That's question number two. And thirdly, on the 5G side, could you share your comments of the market structure? What do you think about the market structure in a 5G world when there are 2 5G licenses and there are 4 players? Could you just comment what kind of market structure are -- is possible, over that 5G time frame?

[ Sachin ], thank you for your 3 questions. Let me take the first one first about stability in consumer revenues for mobility. To be honest with you, we are very focused on ensuring the ARPUs and the high-value customers are retained. We still expect competitive activity. We do know there are a number of MVNOs that are potentially ready to launch. But because of what is going on right now with the coronavirus, most likely that launch has been delayed. But having said that, we're seeing more responsible pricing in the marketplace. We're still seeing lots of different choices. But I'd say it's far too early to say that the market has bottomed out. We are -- we believe that there's no use playing what we call the "seen the volume" game. We believe it's about having the right value and the right customers that generate that value, so we're very careful about managing ARPUs and managing the high-value customers, rather than flooding the market with free SIMs or very low-value SIMs. So we have taken a different approach. But I do believe it will be a little bit more competitive intensity in the mobile market before we say that we see stability for the next few quarters. I'm not sure if Johan wants to add anything more to that?

No, that's very complete, Peter.

The second question about cybersecurity, very fair question. Cybersecurity in Singapore, regionally, globally continues to grow double digit year-after-year. When we announced the acquisition, we were very clear that we see investment to grow at least for 3 years. The business itself is about providing consultancy, then providing build and operate and then providing maintenance to customers. And to be able to provide this capability upfront before you realize all the ongoing managed services fees, you have to build up capability. You have to build up intellectual property. You have to spend money in research and development. And if the business as well, which it is, has a regional presence, you have to involve to invest in regional capability as well. It is a growth, fast-growing part of our strategy, so we expect that the guidance we've given you for 2020 incorporates the contribution to the top and bottom line for our business. The other point I want to make about Ensign, 2 points actually. Ensign itself is pure-play, really deep cybersecurity capability, not a monthly subscription model. And there's a very strong appetite from government clients, from very large multinational companies locally and regionally for that type of service. And yes, we're delighted that more investment will be provided into cybersecurity by government and by other -- in other regional markets. So again, in the guidance, we have incorporated a high-growth in revenue and a minimal contribution in terms of bottom line. In our portfolio of cybersecurity, though, we do have 2 companies which is a part of the Ensign group. One is D'Crypt that has a number of local contracts that they're delivering to. And again, their contribution is built into our numbers. So the portfolio of companies we have is one pure service play and the other one cryptography with more predictable revenues and margins, which is a good portfolio. Your third question about the structure of potential 5G market. It will definitely be unfolding. But let's be very, very clear, there will be spectrum given to all 4 MNOs, millimeter wave, which will allow 5G play by 4 operators. The 2 networks are all about the 3.5-gigahertz spectrum, which will be given to 2. And of course, there'll be wholesaling arrangements between the MNOs as well as the MVNOs. So how the final structure will play out remains to be seen. We are very confident about the bid we submitted. We put a lot of emphasis in terms of network resilience, network redundancy, network security as well as a number of our commercial use cases, that should we be fortunate enough to win the network, we will be in a position to launch. Too early to say how the market will evolve. But potentially every MNO and every MVNO will have access to 5G services, wholesale or from their own factories. We believe the sharing model is a smart, efficient model for Singapore where the investment is -- minimum investment is put on the infrastructure side, and that will create more service-based competition. That's what the investment should be made to provide innovative services, not to keep replicating infrastructure across, say, 4 operators. So we believe in the next few months, a decision will be made. We hope to be in one of those companies, and we're going to get on with building the capability and bringing services to market 3.5G next year when the spectrum becomes available. I hope, [ Sachin], I gave you something...

One just follow-up on that. Just a follow-up. On the 5G wholesale pricing, is there some kind of a maximum thing which is mandated by the regulator? Or it is up to the operators to come out with their wholesale pricing model?

At this stage, the wholesale arrangements, again, it's too early to say how they will evolve and it's really up to the regulator. But we do have on the 4G side commercial arrangements. And typically, should the operators not come to an agreement, there's an appeal process back to the regulator. So our preference is to have the same approach, the same regime of commercially negotiating wholesale rates. And hopefully, we'll reach agreements with MNOs and MVNOs as the market evolves. So at this point in time, there's no definitive regime announced by the regulator, but we are in favor of commercial agreements rather than minimum or maximum.

Let's take the next question from Rama.

I have basically one question, and that has to do with your CapEx outlook. Could you talk a little bit about the 6% to 7% guidance on CapEx? What's baked into this in terms of the guidance, in terms of whether you're -- is spend on -- primarily on the 4G networks? Or does it also take into account the 5G network spend? And how could 5G network CapEx could come in? Is it going to be on top of this? And what time frame will it be coming in?

Before I ask Dennis to provide some more color, let me say upfront that the 6% to 7% CapEx is about the existing business, our existing capabilities in the marketplace. It does exclude 5G. We do believe that as we have a very robust and very fast 4.5G network, interesting enough, last quarter, we won the fastest network not by a third-party survey, but that's by users in Singapore voting themselves through an application. So we have a very robust and very fast network providing decent coverage, great coverage, and we meet all the quality of standards from the regulator. So the ongoing investment required in 3 and 4G is minimal from our perspective, and that's why we believe 6% to 7% is sufficient. And of course, that involves funding some other lines of business. 5G CapEx is not included. Should we be in a position to win 1 of the 2 networks, we will come to the market and we will explain the level of investment we're proposing to airplay over the 5 years in a joint venture and potentially the impact on that on our numbers. Dennis, if you'd like to add some more details on the 5 -- on 6% to 7% please?

Sure. So included in the 6% to 7% are the -- would be the CapEx investments, which are on an ongoing basis required to meet the quality of service standards that are imposed on us as an operator by the regulators. So that's included in there. Also included in the 6% to 7% is certain ISO information system transmission CapEx that we've included in there as part of our transmission strategy. Included in that capital expenditure guidance is also the capital expenditure to be incurred by Ensign and D'Crypt because they are consolidated into our numbers. So those are your elements of what's been included into the 6% to 7%.

Thank you, Dennis.

Okay. If I could just follow up. I mean, Peter, since you mentioned the 5G is a joint venture, is it a technical joint venture, that has -- is it already been decided on that format? And does it mean -- how would it be accounted in for -- as far as the revenues as well as expenses from the joint venture are concerned going forward?

Look, I appreciate your question. But to be honest with you, it's too early to talk specifically about the impact of 5G on our business and in joint venture arrangements. We have made a confidential submission to the regulator. We have to respect the time required by IMDA to evaluate all the different proposals from the 3 different parties. And should we be successful, we will come out and spell that. I think it's very speculative at this point in time to talk about arrangements that require full approval from the regulator. So I apologize for not being able to be more specific. But we need, first of all, to get the green light from the regulator. And of course, we have an obligation because the investments potentially over 5 years are material to the business. No question about that. But we don't want to jump. We really need to allay the time for consideration by the regulator. There may be conditions imposed on the joint venture or conditions imposed for all [ OLED ]. There are still a lot of variable components. All of that needs to be settled. And should we be successful, we'll know the conditions and we will come out and explain to you how the JV would work and what sort of investments, the order of magnitude, what type of specific technologies. So we will be very, very specific once IMDA has made an announcement who the 2 network operators will be.

Let's take the next question from Aradhana.

Aradhana, yes.

Aradhana, are you there?

[indiscernible]

Okay. Let's go to the next question from...

Can you hear me? Sorry about that. My question, it relates to your 5G bid. Just what were the sort of considerations that you took into account to decide for 5G with M1 as opposed to going it alone? And my second question is Huawei. Is it one of the vendors you would still keep under consideration as a potential vendor for 5G?

Okay. Aradhana, thank you for your 2 questions. First of all, I think, about 1.5 years ago, we were very explicit when we were interviewed about how we see the market evolving and investments in infrastructure in Singapore. And we at Star have said from day 1, we believe a smart business model is one that we share the infrastructure. And basically, we then, by sharing infrastructure, you're enabling more services-based competition rather than inefficiencies in building overlays of multiple infrastructure for a limited set of customers. So that business model has worked in other countries between competitors, especially in the 4G and 3G environment, and we were able to at least agree in principle that it can work in Singapore with another company. So we're very delighted because there will be significant savings in CapEx if someone does it alone versus sharing the infrastructure. And again, it's too early to share this with you, simply because we don't know if we're going to be 1 of the 2 networks awarded that. But if you do the calculations and then talk to technology partners and vendors, there are significant savings sharing versus building alternative networks. And also let us be very clear, we've seen the industry for the last few years in revenues declining. So again, we're very conscious, at least StarHub is, in terms of the return on capital. So we think it's an optimum -- sharing a network is the optimum model to drive return on capital as high as possible versus building alternative stand-alone networks and then continue to compete on prices, and you may not get the right return. Your second question dealt with Huawei. Our approach is we're not excluding anybody from expressing an interest in the final RFP that we will issue. We respect companies from different backgrounds and different -- the origin of the company is not something we take into account of the final decision. What we take into account is the ability of the company to have the right technology, the ability of the company to deliver the right technology, the ability of the company to support us, the cost envelope, the investment we have to make, and over the lifetime of the technology, is the road map there for new features, new functionality. That is the most critical aspect of our evaluation. And of course, all these companies have to meet very stringent criteria that had been imposed on the 2 5G network operators for redundancy, resilience and security. So at this stage, we're not excluding any particular company. It will be a very competitive process. It does involve significant investments. So we look forward to receiving bids from any operator who wants to bid for our network when we go to market.

Before we get to the next question, we have a question coming from the webcast. Foong from CIMB has 3 questions. The first would be if management could break down the service revenue growth guidance of 1% to 3% into the respective business segments. The second question is on the dividend guidance. He is asking if that is a firm commitment. Or is that still dependent on earnings performance in FY 2020? And the third question is on the joint bid with M1 on the 5G license. He is asking what is the StarHub-M1 edge over other players in securing 1 of the 2 licenses. And do we see this partnership with M1 also catalyzing deeper collaboration across the whole network?

Thank you very much for the questions. I'll take the third one first, and then I'll ask Dennis to address the first 2. In terms of the competitive edge between the bid we submitted to the IMDA, again, please excuse me for not elaborating on that. I did allude to in the previous answer that it's a more efficient use of capital because it will provide coverage and capacity across the Singapore landscape by 1 investment instead of 2 investments. So apart from the efficiency of the capital that we will deploy, I think it's too early to talk about the uniqueness of our proposal vis-à-vis the other competitors. And again, we don't even know what the other competing bids are. So it plays -- allows us the opportunity for the process to be concluded, and then we'll be quite happy to share with you how we will roll out a network and where we believe is the competitive advantage. It's a little bit too early. Two questions, one on dividend sustainability, I think, Dennis, and the other one, breaking down the revenue guidance.

Okay. So on the 2 questions, I'll take the first one on the service revenue guidance of 1% to 3% growth over the FY '19 service revenue. It really is -- we don't break it down in terms of the guidance However, I will guide that it does take into consideration a number of projects or new initiatives that will come on-stream in terms of the cybersecurity bucket that we've actually invested in last -- in Ensign as well as in D'Crypt. We have also taken into account the structural changes and the competitive landscape in each of our base businesses in Mobile, in Broadband and TV, and the exit ARPUs and subscriber base that we have in each of these lines of business as well as the Enterprise Business in both the managed services that has -- continues to register some growth as well as the pricing landscape on the connected fleet side of the Enterprise Business. So that has all been built into the service revenue guidance in totality. We have also taken into account some of the impacts that we do foresee in the current assessment of the impact of the COVID-19 virus into this guidance as well. On the dividend, barring any unforeseen circumstances, we do intend to pay the $0.09. Given that we are going towards the semi-annual reporting regime as guided by the SGX, which took effect on February 7, we will be paying this on a semi-annual basis starting from this year. So unless circumstances change dramatically, we intend to keep to this $0.09.

And just to add to that, we always have the same communication language about our intentions. And I think in the past, we have been fairly honorable in implementing our intentions. And the last point I was going to make, just to clarify, Dennis, the biannual payment, of course, does not mean we will report to the market only twice a year. We do intend and in fact, we will continue to report quarterly, but in an abridged version.

Yes, absolutely.

Thank you, gentlemen. Let's take the next question on the call from Piyush.

A few questions. Firstly, on the Pay TV, ARPU has sequentially increased in the fourth quarter. So can you help us understand what's driving that? And what would be the outlook? Secondly, on the postpaid ARPU, also that has inched up in the fourth quarter. So is that the exit ARPU trend also? And the outlook around the postpaid ARPU side, if you can elaborate? And lastly, how big is the roaming revenue component as a percentage of Mobile revenue?

Okay. Piyush, thank you for your questions. Great opportunity to hear from Johan, who heads up our consumer business. So Johan?

Piyush, thanks for the question. On the Pay TV ARPU, it's stabilizing after the cable-to-fiber migration. It's, I would say, constant business. We see good uptake on the new introduced plans. So Q3 to Q4, as you said, looks stable. So our expectation is that we manage it to keep that way. On the postpaid ARPU, similar. I think that was a question earlier actually in the queue on that. We're encouraged to see some stabilization. Obviously, this will be depending on the market going forward and the price competition. We have seen some interesting developments in the market where we see some more, I would say, reasonable pricing being deployed. So we're hopeful that if the trend continues, we can stick to that as well. In terms of the roaming revenue percentage, that's something, unfortunately, I can't disclose. It's, I would say, a level which we would like to -- it is actually collapsed in both pre and postpaid revenues in total, so that's where you'll find it.

Okay. And Piyush, again, just a couple of additional comments on the Pay TV. Right now, the base has entirely migrated from cable to fiber. If you look at both companies and some of our quarter 4 numbers were delayed cancellations because due to processing --

Correct.

But those customers have really left us in quarter 3. They had indicated they didn't want to migrate. So we know that the market is definitely not growing. And I think even the alternative company that provides Pay TV did not show any growth in customers in quarter 4. So we expect -- this is an expectation of some rational pricing in linear markets for Pay TV. So if the market behaved rationally, we expect stronger probability for ARPUs to stay stable. The challenge, I think, we all have in Pay TV is not the 2 alternative options. There is piracy, significant piracy. And we've done in the short term, feel confident that's going to disappear. And of course, it's flight to OTT service providers. But now that a lot of our customers, of course, through the migration around long-term contracts, so the number of new customers in the Pay TV, potential market to switch are very few every month. And we just hope that with the rational pricing and only few customers coming to the market, we should be expecting that stabilization. But again, the market behaves the way it wants to.

Sure. Just a follow-up on the Mobile side. Could you elaborate a bit on what have been the latest positive developments on the market side?

Okay. There's a bit of a mixed bag, if I may put it that way. But I think the positive is that, in case you haven't noticed it, we actually priced that to some of our tariff plans a couple of weeks ago. And that's something we have actually been seeing not only with us, but across the market. So I would say that's hopeful. The other thing which we find encouraging is that after a quite eventful 2019 in terms of subsidies in handset pricing, that seems to be stabilizing as well. So that is an encouraging sign when it comes to managing handset subsidies going forward.

Let's take the next question from Varun on the call.

I've got 3 questions. I want to just understand a little bit on the EBITDA margin guidance. So if you look at your operating cost this quarter, there is also one-off included in the -- in one of your line items around, I think, $11 million. And you had in this quarter, some of, again, one-off costs related to migration of subs to fiber. Plus this year, you should have some savings coming from such shutdown of HFC network, the lease center that you're paying to SingTel, right? So based on that, I just -- and obviously, there are cost-saving initiatives that you're doing. Doesn't it look like your margin guidance is pretty conservative? Or is there any significant cost increase that you're expecting in the core business, excluding the cybersecurity? Are you making a lot more investment in cybersecurity where the margins are deteriorating further? That is one. Number two, on 5G, when do you expect to launch commercial services? Hoping that you guys, you're one of the winners. Is it realistic, in your internal estimates, when do you think there will be commercial launch of services? And what is the use case that you're looking at? Is it, again, that you're seeing across the region more mobile-centric use cases? That will be helpful. And again, would you try to push these services at a premium to 4G services? Some clarity on that will be helpful. And lastly, if you can share, in your joint arrangement with NBN, is it both excess and core? Or is there any -- or just one of the network? Or just the excess network?

I will refer the question on EBITDA guidance and the one-offs to Dennis, and I'll address your other 2 questions, Varun, in the next few minutes.

Okay. Varun, on the EBITDA margin guidance, we've actually factored in obviously the -- what we're seeing in terms of the ARPUs going forward on our base business. Bearing in mind, as Peter has guided when he walked through the initial results announcement, the mix of revenues going forward is changing. There are structural changes in our business and the competitive landscapes around our connectivity business and an increase in managed services as well as cybersecurity revenues as well that come into play, which have a relatively lower margin compared to our Mobile line of business. So that's been factored as well. To your question on the one-offs in Q4, we've taken a charge for a cable system that we no longer use as well as certain restructuring costs as well in that. We've also taken into consideration certain IT investment costs for transformation into our EBITDA margin guidance as well. So those have all been factored in where we guided to this margin level.

In relation to the 5G rollout of services and use cases, again, at this stage as guidance is what is happening in various other markets overseas, we see the consumer market to -- surprise in a way, but rationally, it is the right trend. The consumers are jumping first to 5G simply because the availability of handsets is becoming increasingly a wider range. The price points of the handsets are the same price point as 4G. So again, it will be more consumer-led. And also if we take into account that a significant number of customers in the Singaporean market are postpaid with a bundle, with a handset. So as customers coming out of contract every month, they'll have the opportunity to enter into a 5G because if the price for 4G handset is very close to 5G, that will take the opportunity. You have a much faster way of leveraging the network, plus you will always fall back to 4G if initially the coverage is not as widely acceptable as it can be. And every new network will not provide 100% coverage from day 1. So you'll be able to fall back. So that process means that consumer will lead. In terms of timing, again, should we win, the 3.5-gigahertz frequency becomes available first quarter '21. And they're already up to 3.5 gigahertz services is scheduled at that point in time, subject to getting approvals and passing various certifications and network resilience tests and so on. Whether it will be fixed wireless access, whether it will be mobile-centric, I think these are the trends. Now FWA in Singapore, it is less likely, but still possible, especially in new areas, and we're conducting some trials about that. But it will be predominantly mobile-centric. And at some point in time, the enterprise use cases will increase. And keeping in mind that we can serve enterprise with millimeter wave as well as 3.5 gigahertz. So I think the launch of the new -- 2 new license will be towards the beginning of quarter 1 of 2021. We have to build network, change antennas, reconfigure radio as well as build a new course, an [ SA ] course, so that takes a little bit of time. So quarter 1, if everything falls into place, is really the due date. In terms of millimeter wave for specific customer use cases, that can sort of happen some time this year, but it's not going to move the dial at all. I think the enterprise customers are doing more planning. They have to retool, they have to change and they'll take a little bit more time. And we're worried that the scale of potential use cases with the enterprise will not be grand scale. However, millimeter wave is a tremendous alternative to WiFi, secure, fast. So again, we expect, over time, a number of WiFi-based enterprise networks to move to millimeter wave than 3.5 gigahertz. Your fourth question was about what exactly we're doing together with the other company in terms of the joint bid. Again, it's far too early to tell you if it's access and core and backhaul. All of these things will be revealed if and when. And we like to believe if -- or sorry, when the regulator decides. So please be patient on that. But we think it's a good network-sharing arrangement for both companies. The only thing we can say to you, of course, the marketing will happen from both companies. The factory, the network infrastructure, is predominantly the shared type of investment, not the retailing and the branding and the commercial offers to consumers to wholesale. All of that will happen within each entity, not in the joint venture.

This is helpful. Dennis, just a follow-up, if you can give some -- if you can share some number, amount of cost saving with shutting down of HFC network or the least enter that you're going to pay that you're paying to SingTel.

We've never disclosed that number separately, and we respect the confidentiality of the contractual arrangement that we had with SingTel. So even though that obligation no longer exists, we do respect the confidentiality of that and the commercial sensitivities around that. So we -- it is a substantial number. We have said that before. But that's all I can say at this moment.

In the guidance we've given you, we've allowed for the impact of that.

Sure. Is it in the fourth quarter number or you yet to show in the number?

Well, everything that would have been recorded in respect of that arrangement with -- on the cable would have been recorded in 2019. There's nothing left for 2020.

[Operator Instructions] In the meantime, we move on to a question from the webcast. Paul Chew from Phillip Securities is asking, can you -- can management elaborate more on some of the projects that is driving the growth for cybersecurity.

Again, I did mention that the cybersecurity business is built on 4 specific drivers for growth. One is consultancy. One is building the capability for a company. One is operating it, and then maintaining it. That's what drives the revenue. The other part -- the other dimension is multiple dimensions, multiple vectors. Of course new clients, to provide the full range of subservices or new clients for either systems integration or consultancy for maintenance. And the third dimension for growth from cybersecurity is diversification in regional markets, not just Singapore. Because despite how much growth there is in Singapore, and of course, that's the origins of the company, there's still bigger opportunities in regional adjacent markets. So these are the engines for growth for cybersecurity.

Thank you, Peter. There are currently no questions in queue. [Operator Instructions] We'll take one last question from Paul.

So can I just -- I understand -- for the cybersecurity question, yes, I understand the types of -- the general description of the types of what you're doing. I'm just wondering, could you just, if possible, elaborate on a typical project, a typical plain-vanilla project that you do for cybersecurity? And my second question would be, I just wondered, there was a slight in -- a decline in your Pay TV subscribers quarter-on-quarter. I thought most of them were on 2-year contracts. So just wondering the reason for the quarter-on-quarter decline.

Maybe we can start with a question on Pay TV first. The reason why there is a small decline in Q4 is what Peter also referred to earlier on, spillover at the request of some customers to be disconnected in Q4 rather than in Q3. So that's the reason basically.

Look, I wish I could -- thank you for your questions, Paul. But I wish I can say there's vanilla flavor opportunities in cybersecurity because the vanilla flavor predominantly come from the licensing of firewalls and those sort of products to protect clients. Our cybersecurity company is what we call a pure-play, deep. They're doing really threat analysis. They're doing potential offensive scenarios as well as defensive -- building defensive capability. They're doing vulnerability tests. So there is no such thing as a typical assignment because some of the clients may have some capability, and they want to test that capability and build more resilience. Other clients don't really know where to start from. So there is a complete threat analysis from day 1, hypothetical, and then building and testing. So there's no such thing as a vanilla flavor. And what we're seeing, we're actually seeing all of these components, which basically really break into consultancy, build and operate systems integration and then management, all of them are growing. And what also underpins that growth from Ensign, they have a unique research lab. So they're doing their own research, their own innovation and they're building unique intellectual property to be able to develop an advanced security, cybersecurity service, to some of the government and multinational clients. I think I will not do justice to try and explain simply because their level of expertise is unique, and they're doing some of the complex work that other cybersecurities are really not yet able to live to that expectation. So no vanilla flavor, it's more customized based on consultancy more than anything else. Of course, they're running a SOC, security operation center, for maintenance and reporting and so on. But that's a sort of typical operation. But again, within the SOC, they have built in some further intelligent capability using data analytics, using AI. So it's really not something we sort of talk about publicly. They really talk about to specific clients about specific capabilities. It's not vanilla flavor. It's pure and deep cybersecurity play from the company.

Just wanted some elaboration. So just one quick follow-up. In the fourth quarter, I just want to clarify, I didn't really go through the numbers that, in frank, data. But you mentioned there was a fourth quarter charge for the cable system. Is it the $10.9 million that's in the results? I just wanted to clarify that.

That's correct. Yes, that is correct.

But Paul, we issued the pack late. So we understand we didn't give you enough time to go through all the numbers. We apologize. But yes, Dennis just confirmed that.

No, no, that's fine. Thanks for explaining. So that's inside the depreciation and amortization? Because I couldn't see a separate line item apart from the commentary.

No, that is actually included in the operating -- in the repair and maintenance line. There's a separate line for -- in the depreciation, amortization for accelerated depreciation on certain IT systems that we intend to decommission as a result of the IT transformation work that we're undertaking in 2020. There are 2 separate amounts.

Sure, sure. Okay. So the $10.9 million, just to confirm, is on the repair and maintenance part, okay?

That's correct.

And just one final -- sorry, to track this one, but just one last question, on -- could you touch a bit more on any new developments on the content cost that you mentioned in the past to keep it more variable? That's just my last question.

Okay. We're committed to continuing to drive all the negotiations we have with the content providers at the end of that contract. We're committed to drive both the absolute cost down, but change the structure of that cost into a variable cost as well, totally variable. Again, at this stage, we're not a privy to share with you all the contracts that are coming up. But Johan and his team continually drive that. And as we're driving the cost of the existing content down, we're also looking at alternative content that has a variable cost and introducing that to refresh some of our libraries. I think that the only thing we can give you assurance is that we will not accept any contracts upon renewal on a fixed cost as per the past. If the content providers are not willing to show flexibility and introduce a revenue-based model, a revenue-sharing model, variable as well as a minimum sort of minimum guarantee, then unfortunately, we'll probably be making some hard decisions and seeking alternative supply. But that's our approach to the business model to make sure the Pay TV improves always its cash flow and its margins. I don't think we can get more specific than this provided.

Okay. So I just wanted to confirm that there's still runway to lower the cost. I mean there is just 2 contracts yet to be negotiated, yes.

There's always, at any month, contracts. They are the 2-year contracts, 1-year contracts, 3-year contracts. Look, if any material contract is up for renegotiation, subject to the outcome of that, we'll inform the market. But it's business as usual in terms of contract renewals and renegotiations.

In the interest of time, we'll take our very last question for the evening from Piyush.

Just one question. On the dividends, would it be fair to assume that annually we will get a dollar guidance ahead of the year based on your assessment of the profits? Or would you at some point of time change to variable 80% of net profit in terms of dividend payout policy?

Okay. We have adopted a dividend policy in 2019, which effectively says that the Board will pay at least 80% of the net profit after tax attributable to shareholders, excluding any one-off and nonrecurring items. The intention is to go to that policy. We decided to guide the market to an absolute dollar number in terms of a per-share basis this year, but the intention is to actually move towards that. As to whether in 2021, we will still give an absolute dollar number or stick to that policy, it is -- it will be at the discretion of the Board, but we decided to do this for 2020.

Amelia, unless there are any more, I think we've kept everybody at least on for an hour.

Yes. Thank you, everybody, for joining us this evening. This is the end of our call. As always, please feel free to reach out to us if you have further questions. To our next results call. Have a lovely evening.