Good morning, ladies and gentlemen. Welcome to XL Axiata's earnings conference call for the first nine months of 2018. My name is Jeff, and I will be your coordinator today. During the presentation, all participants are in a listen-only mode. Instruction will be given on how to register your question when we get to the question and answer session. As a reminder, this conference is being recorded for replay purposes. Now, we would like to turn the conference over to our host, Mr. Indar. Please proceed.
Thank you. Good morning, everyone, welcome to the call. On behalf of the XL management team, I would like to thank all of you for taking the time to join us today. With us on the call today, we have Ibu Dian, our Chief Executive Officer, Pak Adlan, our Chief Financial Officer, and Pak Allan, our Chief Commercial Officer. Ibu Dian will share the highlights for the first nine months of 2018, which will then be followed by a Q&A session. I will now hand the call over to Ibu Dian.
Thank you, Indar, good morning, everyone. We are pleased to report that this quarter we have seen a pickup in revenue growth, where our gross revenue managed to increase 6% Q on Q. This is a direct result of our successful data monetization efforts as part of our overall strategy. Thus, despite the challenges during the first half of 2018, with intense data price competition and structural changes of prepaid SIM registration, we see that market conditions are improving. If this continues, we are confident that we can finish the year on a strong note. The positive performance this quarter is attributed mainly to consistent execution of our strategy throughout our transformation journey, this has enabled us to gain traction in the market and particularly in winning data-savvy customers.
Further, as we have executed on our plans to monetize data since mid-May, coupled with our strong product strategy and continued network investment program, we continue to outperform the industry in the first nine months of the year. This quarter, our gross revenue increased 6% Q on Q. This was mainly due to continued growth in data revenue, which remains our main growth driver. Innovative data offerings and improvement in data service quality have successfully pushed data revenue growth in the nine months of the year to 14% year on year. Data revenue contribution now makes up the majority of our service revenue at 80% in this current quarter. This percentage is far higher than our peers, which enable us to weather the negative effects of declining legacy services of voice and SMS far better than others.
Positively, our EBITDA has also increased by 9% QoQ this quarter, while margins have also added 1% to 37%. Our continuous effort in driving cost savings have started to show positive results, with total operating costs for nine months of 2018 remaining flat despite the increase in the number of new network sites as we expand our coverage at Java and additional network elements installed to improve network quality during the year. This cost efficiency program will continue and will be a key catalyst to drive higher EBITDA and margins going forward. Data continues to be our main engine of growth. Indonesia's data-savvy customers have continued to respond well to our improved network. As of first nine months of 2018, our smartphone subscribers stand at 42 million, a 15% growth compared to the same period last year.
This makes up approximately 78% of our subs base, which is significantly higher than the industry average. We continue to see a fast rate of migration to subs to 4G, where 4G customers now make up more than half of our total subs base. As more of our base moves to 4G, we are seeing lower utilization of 2G, which will enable us to free up even more spectrum for 4G. This puts us in a better position than our peers and will enable us to continue to best position to ride the growth in data on our journey to become a data-centric company.
We are pleased to report that our customer numbers post the prepaid SIM registration implementation have continued to rise while ARPU remains stable, which is due to our focus since 2015 on sustainable customer acquisition. Thus, we have added another 1 million customers, taking our overall base of registered customers to 54 million this quarter. In establishing the XL brand as the choice for high-value customers, both reliable high-speed data and superior network quality are essential. As such, we continue to ensure a high-quality data experience to our customers through continued rollout and upgrade of our network. Thus, our total BTS count is now over 116,000 BTS, with 3G totaling more than 50,000 and our 4G LTE service is now available in 387 cities and areas across Indonesia with more than 28,000 4G BTS.
To solidify XL Axiata's position as the leader in data innovation, this year saw the first outdoor 5G and WiGig live trial in Kota Tua Jakarta. 5G and WiGig have the potential to enable a variety of services, including enhanced mobile broadband, smart cities, smart home broadband, and digital transformation for vertical industries. We also continue to invest in transmission, backhaul, and network modernization to support the rising data traffic across our network and deliver stability, expand our network capacity, and improve quality of our data services for our customers. Our network investment continues not only within Java, but with a greater focus on ex-Java this year following the strong performance last year, which has translated to better coverage and network performance in these areas. This has also translated to a stronger revenue performance outside Java, which continues to grow at a better rate than Java.
The dual brand strategy has successfully led to XL Axiata, making strong inroads in different segments through innovative offerings in each segment. During the quarter, we launched a higher spec version of our popular Xtream smartphone called the XTREAM Ultima. This offering has done well and is a key part of our strategy to continue to drive smartphone adoption across our base of customers. AXIS also continues to do well with the youth segment, as we have expanded our product offering to appeal even more to the gamers and the music lovers. Our postpaid brand, XL Prioritas, also continues to do extremely well, attracting customers through attractive offerings and smartphone rebate programs. Our performance in the first nine months of 2018 was very promising. We are seeing signs of improvement in the market with gradual price increases, both by us and our peers, which is positive for the industry.
This is evident from our strong performance this quarter as we have undertaken monetization of data. With our positive results, strong fundamentals, coupled with our focus on transforming into a data-centric company, we are confident of delivering a strong finish to the year. Nevertheless, we will closely monitor how the market unfolds, which will have a bearing on our overall performance. Taking all this into account and recent developments in the market, our guidance for 2018 is maintained, with revenue to grow above market, EBITDA margin guidance of high 30s, and CapEx spend guidance for 2018 of around IDR 7 trillion, which will remain focused on data network investment in 4G and continuous network improvements and modernization in and outside Java. Thank you. Let us now proceed to the Q&A session.
Thank you. To ask a question, please press star one. To cancel, please press the pound key or hash sign. Please kindly but strictly limit your question to only two and to allow other participants to raise their question. Should you need to ask more questions, you can go back to the queue by pressing star one again. We shall end the conference call sharp at 11:00 A.M. Jakarta time. It's 2:00 P.M. Jakarta time. First question comes from the line of Piyush Choudhury. Your line is now open.
Yeah. Hi, good morning, everyone, and thanks for the call. Congrats on the good set of numbers. Two questions. Firstly, on data tariffs, we saw a further decline in data realization quarter on quarter for XL. Do you think there is room for a hike in data tariffs by XL? Do you think current pricing is reasonable to generate required return on invested capital? Secondly, in ex-Java, could you share your achievements versus objectives and what are your network coverage objectives in ex-Java for 2019 and 2020, and what it means for network rollout and CapEx? Thanks.
Yes. Thank you for the question. Let me take the first one regarding data pricing. If I look at the first 9 months, you're right. In the beginning of the year, we were facing the challenge coming from Q4 last year, we saw a decline in the prices, basically both voice and data in Q1. Since then, here I'm talking around April, May, we have basically been increasing the prices in the market. At the same time, we also see there is room for more. You have to remember that we are not, because it's very different here in Indonesia compared to other countries. We are not in charge of the end user price, as we are selling a price to the dealers, to the distributors, then they can basically do whatever they want.
We see a slight increase in prices in the market regarding data, we basically welcome that. You're absolutely right that we still have room for that going forward now that we see there is some ease after the SIM registration process. Yes, we believe that we will still see some prices increase in data, it will depend on if it's Java or it's ex-Java. Some of the places we still need to be pretty aggressive, especially with ex-Java. To the network ex-Java?
On your second question, we started investing in ex-Java, I would say 18-24 months ago, right? The objective of expanding ex-Java is really to gain market share. The situation is very different when you look at Java and ex-Java. You look at Java, from a network perspective, if you compare the three operators, we are quite similar in terms of network size. Probably the incumbent have probably a slightly bigger network. As of today, we are quite equal from a network perspective. However, from a market share perspective, we probably have one-third, one-third each. Probably now XL and the incumbent probably slightly more given what we are seeing in the market with one of the other competitors.
However, in ex-Java, you probably would also know that Telkomsel have actually a dominant position right now with 80% market share. Pricing outside Java is also extremely expensive. We saw that as an opportunity
I think we have aggressively rolled out outside Java. By end of this year, I think you would probably see that our 4G coverage would be around 80% of population coverage. That gives us ample ammunition to probably compete with the market leader. Since we've done the rollout, I think we have probably seen that ex-Java have been actually growing double digits. I think the trend is actually continuing. We are seeing good traction coming from outside Java. We have also seen that the perception from our customers on network have also improved significantly with the improved network quality outside Java.
Sure. [Allan], on ex-Java, follow up out there. In terms of concrete kind of objectives, could you share what's your network objective only in ex-Java? Because 80% pop coverage is probably for pan-Indonesia, right? What does it mean in terms of CapEx by 2020? Secondly, sorry to follow up on data pricing, but I just wanted to get a sense on, we are still seeing the blended pricing decline, right? As again, as one of your competitors who has reported an improvement in data realization. I wanted to gather your thoughts. Is there a kind of improvements? Is there a massive case where you can increase prices massively in Java or you would continue with your current strategy to gain market share? Thanks.
Let me take the first one just to elaborate on the pricing. I think massive is a very strong word. I don't think we'll see massive price increases in the market for sure. Just to make it clear, we are so happy with the result right now because we are following the game plan. The game plan is basically back to 2015, 2016. We're going to be a data centric company. Looking at the first nine months, we are not only looking quarter by quarter, we are looking for the first nine months. We are very happy with the results, and we are following with steady growth quarter by quarter. Saying that, at the same time, we welcome any price increase. You're right that we see some price hikes in the markets in some specific areas.
We look at my spreadsheet for all the zero price in this market, we have seen increases. Yet we still believe there will be some price increases in the market going forward. I don't think we will call it massive price increases.
Piyush, on the ex-Java, when I said 80% population coverage, 4G population coverage, that is purely referring to ex-Java. We have got quite a decent network coverage by end of this year to be able to compete with market leader. From a CapEx perspective, I think you've probably seen that today we have reallocated some of our CapEx for the investment outside Java. You would probably expect that going into 2019, that allocation will probably be skewed a lot more to outside Java.
Sure. Thanks a lot.
Our next question comes from the line of Sachin Salgaonkar. Your line is now open.
Hi. Thank you for the opportunity. My two questions are, first one is, you know Pak Adlan and given the initial comments where almost half of your customers are already on 4G and you guys are seeing a faster utilization. Is there a plan to shut down 3G completely and use that spectrum up for 4G? Sort of a related question is, you talked about investments outside Java. When we look at an absolute amount of CapEx going into 2019, should that be similar to that of 2018? You see a bit of a decline out there given the fact that you already have 80% pop coverage? That's question one. Question two, sorry, again, to go back on pricing, let me ask it the other way. What will it take you guys to raise tariffs further?
When we generally talk to your competitors and we had a call a day back, they continue to indicate that they have increased the tariffs and the ball perhaps is in your court. Wherein perhaps the premium or the difference between the tariffs is close to around 30%-35% and it all depends upon how XL is looking at tariffs. Generally, they are happy to take it further, but it appears that indication is involved is in your court. Just wanted to understand, how do you look at tariff increase? When could we see the utilization rates or data realizations actually improving for you guys?
Okay. I'll take the first question, Sachin. I think you probably see that, I would say today, more than 70%-75% of our total traffic are probably on 4G. Our traffic on 2G is very small. Hence, I think there is a possibility of probably shifting some of the spectrum used for 2G now towards 4G. The debate whether we're going to shut down 2G or 3G is something that we are probably having at this point in time. Obviously, I think there are various views on which are the technology that we should be retaining and which we should be shutting down. We have not come to a landing yet, but obviously, I think it's something that we are seriously looking at.
At some point in time, I think looking at maybe later this year or early next year, I think a firmer decision would probably be made accordingly. On the investment outside Java, I think, yes, I think we probably achieve about 80% population coverage this year, by end of this year. Nevertheless, I think there's still quite a fair bit to go. I think from an investment, total investment perspective, I think we are still doing our business plan and which I think we'll probably announce in January when we announce our full year results of 2018. I think from what we have probably seen, there are a clear indication that I think it will probably skew a little bit more towards outside Java.
What that number will be, I think would probably depend on the total overall CapEx that will probably be approved by our board and shareholders. The absolute, I think will continue to come back, but I think ratio wise, definitely skew will be towards outside Java.
Coming back to the prices again, sorry, I have to repeat myself again, that we have increased the prices for the last seven, eight months for sure. Maybe if you guys one day come to Indonesia, I will gladly invite you to the market so you can actually see how the hero products work. Now, usually I do not comment on our competitors' results, you mention it, we heard about the call, right? We were also surprised. We heard that, yes, Telkomsel is saying, we are increasing the prices and we are waiting for XL to do the same, which we have done.
We were surprised seeing that Telkomsel have a data traffic increase around 9% and at the same time, they have a revenue coming from data at 18%, even though the whole market has been increasing 80%-90% data traffic. Saying that, we are not actually sure where that data traffic is coming from or the revenue coming from, it could be something about the stock market, sorry, the stock of the data, et cetera. We are surprised, again, we have increased the prices. We are not behind, we again welcome if some of the other guys are increasing the prices, we will also take the lead in some of the areas here in Indonesia when it comes to price hike.
Okay. Thank you.
Next question comes from the line of Wei Shi. Your line is now open.
Thanks very much for the opportunity. First question relates to the prepaid subscriber base. I've noticed that your competitors lost prepaid subscribers during the quarter, whereas XL has actually gained subscribers. I just wanted to check whether this was due to a timing issue. Would we see a delayed impact on XL in the coming quarters? Any sort of lingering impact from the SIM registration, for instance? Would this conversely be related to the increase in discounts that you are offering during the quarter? Were you incentivizing dealers a lot more aggressively than your competitors? Second question is, with the change in management at Indosat, have you seen any increased willingness by the new management around collaboration/any progress in talks about partnerships with them on the network side?
I think on the prepaid stuff, I cannot comment on what happened to our competitors. What we can say that in terms of defining prepaid subscribers for us, we have been super consistent all this while, how we define subscribers, and there is no change in the definition as far as we are concerned. I think the increase in net adds this quarter clearly reflects what's happening in the market, and definitely we have probably gained a subscriber share. I think there is no change in terms of definition per se. On the competitors, I really cannot comment why there's a huge fluctuation in the numbers of their prepaid subscribers. Secondly, you talk about-
Regarding the question was regarding the Indosat new management team. First of all, we welcome these new guys. What we see from them, this is very mature and very experienced guy coming in the management of Indosat. Again, this is very early. They have just started. We know they have been going around in industry, sorry, in the regions to see what's going on. We basically do not know what happened right now. We do not expect to see any irrational decision in the market at the moment. We hope that they will continue the journey that they've been on so far after we had the SIM registration. Again, we welcome the team, and we are looking forward to see what they're doing in the market.
If I can just follow up, has there been any progress or any improvements in terms of the partnership? We've talked about network sharing which stopped for a while. Have you seen any progress or improvements in that situation with the new management team? I appreciate that they're fairly new, have they made any contact or given any indication that they may bring that forward?
We have not made any contact with the new management because they've just recently appointed. We are very welcome if they want to initiate another discussion with us on network sharing on any other form of cooperation.
All right, great. Thank you very much.
Next question comes from the line of Ranjan Sharma. Your line is now open.
Hi, good morning. Thank you for the presentation. It's Ranjan Sharma from J.P. Morgan. A couple of questions from my side. Firstly, regarding your CapEx plan, so first see that you have managed your CapEx
The absolute CapEx amount from 2016 to this year, but effectively you have increased the contribution to ex-Java. You have been emphasizing ex-Java strategy, but what I'm concerned about is your Java network, because effectively you have cut your Java CapEx by a third while your data usage has doubled in the last couple of quarters. Are you not concerned that reducing CapEx at a time when data is exploding within Java is going to impair your network within Java, and that can risk market share losses? I know the network is fine now. I'm looking forward, let's say three quarters or four quarters from now. That's the first question.
Secondly, you talked a lot about your radio equipment and on what you're investing, but there's also a need for fiber and submarine cables outside of Java, and that tends to be expensive, but can lead to bottlenecks if you don't invest in that part of the network. If you can share your thoughts on how you might mitigate some of the challenges as you see ex-Java data usage increasing. Thank you.
I think, if we look at our investment this year, I think the ratio between Java and ex-Java are close to about 50/50, right? To say that we are not investing in Java is also not true. No doubt that, I think, we have reduced slightly the investment within Java, because of the fact, I think, from a population coverage within Java, we've probably arrived at close to about 90% or 91% or so, right? From a coverage perspective, I think we've covered almost of the Java area. I think the bulk of the investment that's going into Java this year is primarily in terms of upgrading capacity as well as investment in the transport, including fiber, right? Next year, I think, the capacity expansion will continue, right?
As I said, the absolute amount is something that we will come back to the market when our plan has been approved in January. No doubt that there will be some skew towards ex-Java. Rest assured, I think network quality is something that is absolutely important to us and to our customers. We will monitor that very closely, right? I think as of today, as you said, there's no indication that the quality of our network have actually deteriorated in Java, and that's something that we will also ensure and monitor as we move into next year as well. Secondly, I think, over the last 12 months or so, I think, we have started investing heavily in terms of fiber. I think there is quite a number of collaboration as well that we are probably doing on the submarine side.
Just to give a feel, I think by next year, we are probably looking at fiberizing close to 10,000 of our sites. I think within the next three years, we are probably looking at more than 50% of our sites are probably fiberized. I think we are adopting a model where it's probably a lot more innovative in terms of how this fiberization project gets funded. I think we are adopting the same business model as how we are doing it with towers. It's more of a long-term leasing model. With that, I think it enables us to scale up in terms of fiberization quite quickly and quite extensively within this one or two years. Secondly, on submarine as well, you probably have seen some announcement as well. There have been quite a number of collaborations that we have probably done.
I think in partnership or something that we've probably done with some of the other submarine owners. The one that is up and running, our collaboration that we probably have with Australia Singapore Cable that connects our submarine from Sydney to Jakarta to Singapore. I think with that route today, we are the only provider in Indonesia, have an alternative route going out of Indonesia and not depending on. I think we see that this new cable that we're having are probably getting huge demand in the market as well. On top of that, there are other plans that we are having, collaboration that we're having with other partners, which we cannot share with you at this point in time.
Rest assured, submarine and fiber are part of a key planning in terms of coming out with a better network quality for us in the years to come.
Okay. Thank you.
Next question comes from the line of Colin McCallum. Your line is now open.
Hi there. Thanks for the opportunity. Just one question from me and one comment. Question is just on the, from what you were saying, Adlan, just on the fiber side. In terms of extending that out to actually addressing the fixed broadband residential or enterprise market, can you just explain what you're thinking in that regard as if there's a business case worth looking at there and what the extent of your plans might be in terms of quantum of CapEx or targets for that sort of area if you do move into it? I was going to ask about data pricing. I won't dwell on it. I think there's been enough questions on it. I guess comment for Alan, who sounds a bit exasperated by all the questions.
I think the issue is that it's always difficult when what's being said in the call isn't just not matched by the numbers, right? We've had one competitor said they've raised prices and the numbers show that they have, and the revenue's gone up. You guys are saying you've raised, but your numbers are suggesting that you haven't. That's why there's so much difficulty on this. I won't dwell on it. If I could get an answer on the fixed broadband, that would be fine. Thank you.
Colin, on the fixed broadband, I think it's something that we are closely looking into it. As we said in earlier calls, we are currently doing a pilot. I think we are addressing not only tier 1 cities but also doing pilot in tier 2 and tier 3 cities, just purely to understand the behavior of a consumer and the opportunities that we see in these cities, right? At this point in time, I think as we are doing our business plan, we have not come to a conclusion yet on this. Rest assured, I think early next year when we announce our CapEx numbers and all that, this is something that we probably will come back to the market.
The plan on fiber, I think, as you know, that as your traffic grows with 5G coming into stream and all that, fiber is actually a key component of the overall network for a better quality network. Hence, I think we are taking a longer-term view in terms of building the fiber upfront. I think we have probably looked at the business case whether to serve through microwave or to go direct to fiber. Right? From a business case perspective or from mid to longer term, it makes absolutely sense to go direct into fiber, right? I think with the new business model that we are adopting, similar to what we're doing with towers, I think that will also allow us to be a little bit more aggressive as it is probably a longer-term leasing arrangement as what we have done with towers.
Yes, I think we have looked at the business case. With the increasing frequency that we are seeing as well being charged by the regulators, it makes absolute sense to go direct to fiber rather than continue to invest in microwave.
Got it. Thank you.
Next question comes from the line of Arthur Pineda. Your line is now open.
Hi. Thanks for the opportunity. Sorry to belabor the point on pricing. Can we just get some clarity with regards to your pricing strategy outside of Java? What kind of discount do you normally apply when you enter these new markets? How sustainable are these discounts, as presumably the economics are far worse outside of Java? Second question I had is, again, on the pricing side, I just can't reconcile your pricing trends. Your RPMB is down 7% QoQ, even though you've mentioned that you've raised prices. You've also mentioned that it's partly because of the distributor issue, you cannot control the pricing down on the distributor level. Why aren't you able to control data yield on the distributor level? Is that not dictated by the telco? Thank you.
Let me take the first one, then Adlan. The pricing strategy outside Java. We cannot reveal exactly what kind of discount we are taking, where we are doing, et cetera. There's no doubt about since many of the places outside Java is a single-player market, meaning that we only have Telkomsel from there. Now that we are building sites, we are building networks, we of course need customer to justify that investment. To get these customers, we need to be pretty aggressive on the pricing on the product setup. In these areas, yes, we are pretty aggressive. At least to start with, that we have the go-to-market plan where we acquire customers based on a very good price, based on a very attractive product. Yes, we are aggressive outside Java.
Arthur, let me try to answer his point on pricing again, right? If you look at first is on the pricing to the dealers, to the consumer. If you look at what we can control is the pricing from dealers selling to the retail outlets. That we actually dictate in terms of pricing that a dealer sells to retail outlet. However, in Indonesia, typically, when a retailer sells the data packet to the consumer, there's always a markup. What Allan was referring at his initial comment that pricing, that markup is something that we are not able to control. Right? Typically, these retail outlets, we are probably talking more of the mom-and-pop shops. The traditional channel itself.
If you talk about the modern channels, you talk about the Alfamart, the Indomaret, or any franchise, for example, or even, let's say, selling at the bank. The modern channel itself is something pricing we dictate and we control, right? In this sense, the part that the operators are still not able to control the end pricing to the consumer is mainly from the traditional channel, right? On the modern channel, yes, we have absolutely full control of that. Obviously, as we expand more and more towards modern, and we shift more and more towards modern, we probably have a better control over the pricing going to end consumer. If you look from a yield perspective, yes, you probably have seen that QoQ yield have declined slightly. Yield have declined slightly.
I think we have actually mentioned this before. Yield in itself is not a clear indication of pricing in the market. I think you probably need to see where your starting position is today as well. For us, obviously, I think from the overall yield, we are probably going to be lower than the industry, given where we are in 4G. You know that pricing for 4G, given the big throughputs and all that, I think we have a much lower yield. At the same time as well, cost of 4G is also much lower than, let's say, 2G or even 3G, right? Given the fact that our smartphone penetration is really at around 80%, that probably has resulted in probably yield to be much lower than the overall industry.
Secondly as well, I think what we have probably seen as well, it's also part of our strategy, we are also pushing our customers to subscribe to a bigger data packet, right? You know from a yield perspective, as people move from a sachet to a big data quota, the yield is also much lower. It's part of our strategy. We want to push customers up towards a big data package as well. That has also partly contributed to that, right? Thirdly as well, I think you probably have seen, especially lately, I think post prepaid registration, we have actually seen that the migration going from 3G to 4G has been quite substantial, right? It goes back to the argument that the yield on 4G is going to be much lower than 3G as well, right?
I think that primarily applies to AXIS, which I think in the past used to be more of a 3G-driven customer. Today, since we started promoting 4G on AXIS, I think we have seen a significant pickup on 4G traffic, especially on AXIS. While I think definitely we have actually raised prices in the market, especially in the Java area. Because of the other factors that what I've mentioned, that's probably putting some pressure on the yield, and that's what you're probably seeing in the numbers.
Understood. Thank you very much.
Next question comes from the line of Prem Jearajasingam. Your line is now open.
Hi. Thank you for the opportunity. Sorry, we're going to go back to data pricing. Could I clarify this, right? Are you saying that the yields are down because you are able to get people to buy bigger and bigger buckets, and those people are potentially consuming more of that bucket, therefore pushing down the yield even further? To top it all off, because of your go-to-market strategy in the ex-Java areas, that potentially, although the price points ex-Java are potentially higher than Java, but because of your go-to-market strategy, potentially that's diluting that number more than it should in the early days. As you take away those promotional numbers, we get some form of a yield support in there. Would that be a right read of what you've been saying?
I suppose the second question is, how afraid are you that given the situation of the market where your competitor says you've not raised prices enough, et cetera. That we go back to a situation that we saw probably 10 years ago, when they suddenly decide, if you want to bleed, let's see how much you can bleed. They decide to cut prices by, say, 50%, 60% next year, and they become the only party that makes money in the market. How significant a risk is that?
Let me take the first one, and thank you for the summary. I think you were spot on, thanks a lot. I don't want to have more comment on that one. Maybe Adlan also want to answer the second one. We have not discussed that risk yet. Of course, that's a risk if they basically what you're saying, if they want to kill us, they can kill us. Again, we are in the same boat and we want to increase. If we see it in the market, the price increase is happening, we accept all that. As you said before, when it have cooled down ex-Java, we again will be able to increase the prices when we get our fair share of the market where we are investing in the network.
Yeah, Prem, I think the risk will always be there, right? We have probably seen earlier this year as well, right? I mean, late last year. Right. Our strategy is from our perspective is pretty clear, right? Java, outside Java. I think rest assured, data monetization I think is something that we will continue to push in Java. I think that's where you probably seen the battle that have been over the last so many quarters that have drive data pricing down, right? That's essentially in Java. I think when we are talking about all these price increases and all that is all coming in Java, right? Primarily in Java. I think generally as an industry as well, I think we have also seen the industry is also pushing price up, especially in Java, right? It's a different story altogether outside Java, right?
You see a very skewed scenario where the incumbent have more than 80% market share, right? Is that a situation that's probably sustainable in the longer term? Probably not. I think you have hardly seen any sort of this scenario taking place in any other markets around the world, right? At some point in time, that 80% will probably need to come down. Essentially, the only reason why that has been for a long time is because there has been a monopoly situation, right? Now, given that there is an alternative network that's present outside Java, I think consumer will also have the choice to choose within the more expensive and incumbent or the alternative provider, right? From our perspective, I think we are present outside Java to gain market share, and I think our pricing strategy will probably reflect that. Right?
In any case, I think we are also absolutely clear that from a price perspective, outside Java, pricing is still higher than within Java. Right? To say that we are going to kill the market by reducing price is also not the intent at all. Right? For us, it's more of a portfolio and I think outside Java have got its own separate objective.
Perfect. Just one follow-up. Do you think the changes at Indosat could mean a more aggressive Indosat even in the near term, or do you think that network differential is sufficient to provide you guys with market share gains without needing to play the price game?
First of all, we are not in a situation where we can comment on what we expect our competitors to do. The only thing we can comment on was an article saying that they have got X amount of million U.S. dollars which they will spend. Meaning that, yes, it could happen that due to more network from Indosat as well, we could see increasing competition. That's the only thing we can comment on.
All right. Thank you very much and good luck.
Next question comes from the line of Chong Chen Foong. Your line is now open.
Hi. Thanks for the call. Two questions from me. Firstly, on the guidance for EBITDA margin of high 30s, which you have maintained. I'm just wondering whether you think that that could be a bit difficult to achieve because your margins must go north of 40% for you to get to high 30s for the full year. If you do think that will be the case, which cost lines do you expect to ease off further into the fourth quarter? Would most of that margin expansion come more from revenue improvements? On to the second question, regarding your network managed services contract with Huawei that's coming to an end of that seven-year period this year. I just want to understand what will happen next, and should we expect possibly higher or lower costs going forward on that portion of your cost items? Yeah. Thank you.
Thank you, Foong. I think if you look at the network guidance, we have maintained the high 30s. I think looking at our cost exercise that we have actually done, we know that there are certain cost items that would probably be realized sometime in quarter four, right? We are quite optimistic in terms of achieving these numbers. I think where is it going to come? It's a combination, right? Definitely we are looking at from the revenue line moving into Q4. I think if the trajectory, the momentum of what we think in the market, we expect Q4 would continue to be a good quarter as well, right? The margin guidance stays as is, and I think we are expecting this to be achieved from increase in revenue as well as some realization on the cost optimization that we've been working on so far.
On DMS, it's not expiring this year. It will expire next year. Right? Obviously, I think we are in a tender situation. I think by end of this year, we're probably not going to be there, but I think it's something that we are retendering, and I think we need to look at how best to make sure that we get an effective overall pricing from the new tender that we are probably doing. Right? It's still too early to comment on that at this point in time.
Okay. Adlan, just a follow-up on the margin side. If we look at the next few years, right, do you think that XL can keep the overall or absolute cost rather sluggish? I think you've done a really good job the last few years keeping it flat, but the next three years, do you think that's possible, or do we see quite a fair bit of upward pressure?
I think, Foong, the answer, obviously, I think there has been a challenge from our BOC as well of stakeholders, right, to probably keep OPEX flat, right? We know that we have quite a rigorous cost program that we are running within XL to look at every cost line item to see how we make structural changes to make sure that it's sustainable in the longer term, right? I think at this point in time, what I can say is, yes, I think it's something that we are probably looking at. Whether we are able to do that or not is something that we probably need to see.
Rest assured, what we can say is that even if there is pressure on cost and cost increasing, as a percentage of revenue, cost as a percentage of revenue, we expect that to improve as we move forward into subsequent years. Yeah. Whether we are able to keep OPEX flat is something that probably is still a challenge. It's something that we are working at, but I cannot tell you at this point in time whether that's really possible or not. As a percentage of revenue, it's something that we are probably more confident of.
Okay, got it. Thank you so much, Adlan.
Next question comes from the line of Krishnakumar Srinivasan. Your line is now open.
Hi, Ibu Dian and Adlin. Thanks for the opportunity. Three questions from me. My first question related to the previous question on trade channels. Can I get the latest check on your latest distribution channel mix, the split between general trade and modern trade? I recall it used to be 60% GT and 40% MT, but I suppose your GT portion may have increased now as you expand your trade channels in ex-Java too. Perhaps on your comparison on your retail touchpoints in ex-Java. Is it going in line with your network coverage there as well? My second question is on the fast rising bundling cost. I know this is in line with your fast-growing bundling revenues too. We saw IDR 200 billion of bundling cost in the quarter alone. Should we expect this quarterly bundling cost run rate to go higher in the coming quarters?
Can we get some idea on the traction of this bundling program, especially in terms of subscriber addition? I understand if you don't have the bundling subs addition data now, I'm happy to take the figure offline too. My last question is on the internal upselling trends. Now that industry churn rate is somewhat stabilizing, do you have an estimate on how much revenue growth potential that XL can get internally by upselling your internal customers? Say from moving your customers from AXIS subscription to XL subscription. Do you have any data or evidence that you could share to suggest that internal upselling is happening at XL? That's all. Thank you.
A little bit difficult to hear your question number 1, as I understand, it was about the channel outside Java, is that correct? Our strategy regarding channel outside Java, is that correct understood?
That's right, yeah. On your comparison, how fast are you growing your trade channels in ex-Java?
Yes. What we have, we have a project we call the go-to-market project, meaning that every time we put a site somewhere outside Java, we have a very clear KPIs for our go-to-market team. How many salespeople do we need to have? How many more pop shops? How many modern trade, et cetera. I cannot reveal how many it is because that's very confidential. At least we are following that plan. For us, it's one of the most important KPIs that we have the distribution and a channel outside Java. Not saying it's easy, it's very promising so far the way we are actually delivering on our KPI for the channel. We are actually following what we have done here in Java. It's working like that.
Yes. On the bundling cost, yes, I think we have been pushing 4G smartphones device bundling quite aggressively starting from Q2 onwards, post prepaid registration. You probably see this quarter that I think we are probably doing around IDR 200 billion of bundling device. What we can say that we are seeing very good traction. We are probably doing close to 2,000 to 3,000 a day of registration. I think we would expect that trend will continue. So I think this quarter the bundling cost is around IDR 200 billion. You should expect that I think moving into subsequent quarters it will be around that number or probably slightly more. I think this is a key part of our overall strategy as well as we move to become a more data centric company and pushing our subscriber up towards 4G. Third question on internal upselling.
I think we are doing this quite granular at this moment. With our CLM process and all that where today we are selling to all our existing customer where we actually have a platform today that enables us to communicate with our customer regardless of the various channels that we have. Whether it's SMS, whether it's app, whether it's through web and all that. A single channel of communication to the customer that plan to actually upsell all our services. This is going to be a key part of our strategy as we move towards more detail in terms of data analytics and things. So far, the engagement has been quite good.
I think as you probably would expect as well, at this point in time we are able to give personalized offer to each and every one of our customers depending on their behavior and usage pattern.
All right. Is there any tangible data or any evidence on if you're seeing any ARPU upgrade from the upselling initiatives so far?
I think we definitely see an uplift, but I think that info is probably not something that we probably want to share to the public.
Sure, that's fine. Thanks. Very helpful. Thank you so much.
Next question comes from the line of Gopalkrishnan Kumar. Your line is now open.
Yeah, hi. Thanks for the opportunity. A few questions. Firstly, there's been a fair bit of discussion on this price increase. My question is on profitability. You continue to report losses because of your higher D&A CapEx related to D&A and debt related charges. When do you think XL should see a bit more sustainable profitability? Is it still a few quarters away or do you have a firmer confidence of this happening a lot sooner? It has linked to the revenue growth as well. That's my question on profitability. Second is on the ex-Java business. What's your market share now versus any target that you can share in this market?
How many more quarters of this pricing strategy do you think you'd have to continue to reach your target? A related question is on the interconnect cost side, which seems to have gone up quite sharply sequentially. Would you expect this cost trend to continue on the interconnect cost? Last question is on the payable side. I see that the third-party payables in the balance sheet have gone up and it's now around IDR 8.3 trillion. Can you talk a bit on what's driving this and the outlook here? Thank you.
On profitability, yes, I think we see quite a fair bit of pressure on the profits this quarter, especially given the fact that I think we are building a lot of new sites. Post-Lebaran, I think we've seen that quite a number of new sites have actually come up. I think you probably see some impact of the lease rental coming up in quarter three per se, right? I think there's a few drivers that we are probably looking at in terms of driving profitability. No doubt, as we continue to invest, D&A will increase, right? One point that I probably want to see that a lot of these sites are probably just being put up.
I think what we are probably monitoring is to make sure that we build these sites that actually been built as quickly as possible. We are monitoring site by site. We are looking at revenue per tower. We are looking at profitability per tower. I think this tracking is done on a very regular basis, right? For each and every one of our retail people, they are probably aware of the target that has been set in terms of loading up the revenue and per tower per se in each of these new clusters, right? Growing revenue is absolutely key in terms of driving up profitability. Having said that, I think not just from a revenue line, we are also looking at each and every one of our cost items.
I think you probably have seen that as much as possible, we are trying to keep our OPEX flat, which we have been quite successful in doing that this year. I think that exercise is probably going to continue next year, right? I think we are probably not far from achieving profitability on a quarter basis. I think we'll probably see as well that I think if we are able to continue to grow revenue in quarter four and looking at some realization of the cost element, we should be able to end in quarter four in a profit situation, right? Having said that, I think one of the other key unknown factors is probably on Forex, right? I think the impact on Forex, I think the debts are all hedged up to maturity. However, I think one of the bigger impact is probably coming from CapEx.
As much as possible, I think we are trying to push our CapEx from foreign more to IDR, and we are doing a lot of work on that to see how we can do that, right? To answer your question, going to profitability, revenue is key. Looking at all cost elements, I think it's something that we are already doing now. I think we are probably not far in terms of achieving that profit numbers in subsequent quarters, right? At Java-
Market share in East Java, right? Not to share because we don't have the exact figures, but as we know that almost 80% is coming from our incumbent, meaning Telkomsel. We can say we would mainly have a fair share of the remaining 20%. It's more important as you ask how long time will this continue, and we right now see at least for the next four quarters, we will continue this with the investment, with the go to market, with hiring people in these markets, and with hiring new retailer and new outlets. That will at least continue for the next four quarters. The last question was?
On payable, I think yes, you've probably seen payables have gone up this quarter. You've seen that probably paid CapEx is also down this year. I think this is primarily attributed to some vendor financing that we have secured with some of our key vendors.
Thanks. Just two follow-up questions. How much of the CapEx is in local currency versus Forex? When are you expecting to lower these payables? You talked a bit on the Huawei side earlier.
These payables, whatever the case may be, the payables first is interest-free, right? It's more of a longer cycle of a supplier's credit, right? As and when we have actually renegotiated for a longer credit period, right? I think as we go along, for example, on all these new purchases with some of our major vendors, we would probably have a longer credit period. What's the other one? Sorry. What's the other one?
Sorry. My question was how much of the CapEx is local currency?
Local currency. At this point in time, we have USD CapEx is approximately about 45%. On the overall CapEx is in USD. There is a plan to try and renegotiate this down, right? How successful is it going to be is something that we need to see. Let's see sometime next year whether we are able to drive this USD purchase down, right? Thirdly, there's a point on you asked about interconnection as well, right? That's interesting, right? When we see in quarter three, you see a sharp reversal of the interconnection. In quarter two also, we were in a slightly net in position. We have moved to a net out position. This is when we actually analyze the numbers, it's probably coming from the market leader, right?
What we have seen is they have increased price, significant increase in price for voice, especially off- net. That's probably drive this behavior from a consumer perspective, right? Two things will happen, right? We have seen this happening as well in the past, we have gone through that cycle as well, right? First, you probably see a change in consumer behavior. What you could expect as well that given that the price increase is quite steep, quite substantial, you probably would see some of these customers would probably churning out as well, right? Anyway, we are monitoring this situation. We are also seeing some impact on the interconnect, but definitely some positive impact in the market.
Okay. Thank you and all the best.
And now-
Okay.
Back to the call to your host. Please proceed.
Thank you, Jeff. Thank you everyone for your participation in today's call. As always, do get back to us if you need further information. We'll see you again next year. Thank you.
That concludes today's conference call. All lines, you may now disconnect.