Bharti Airtel Ltd
NSE:BHARTIARTL
If you don’t study any companies, you have the same success buying stocks as you do in a poker game if you bet without looking at your cards.
Utilize notes to systematically review your investment decisions. By reflecting on past outcomes, you can discern effective strategies and identify those that underperformed. This continuous feedback loop enables you to adapt and refine your approach, optimizing for future success.
Each note serves as a learning point, offering insights into your decision-making processes. Over time, you'll accumulate a personalized database of knowledge, enhancing your ability to make informed decisions quickly and effectively.
With a comprehensive record of your investment history at your fingertips, you can compare current opportunities against past experiences. This not only bolsters your confidence but also ensures that each decision is grounded in a well-documented rationale.
Do you really want to delete this note?
This action cannot be undone.
| 52 Week Range |
971.55
1 771.1
|
| Price Target |
|
We'll email you a reminder when the closing price reaches INR.
Choose the stock you wish to monitor with a price alert.
This alert will be permanently deleted.
Bharti Airtel Ltd
You don't have any saved screeners yet
Ladies and gentlemen, I'm Rajyita, the moderator for this webinar. Welcome to the Bharti Airtel Limited First Quarter ended June 30, 2022 earnings webinar. Presenting us today is the senior leadership team of Bharti Airtel Limited. I must remind you that the overview and discussions today may include certain forward-looking statements that must be viewed in conjunction with the risks that we face. Post the management opening remarks, we will open up for an interactive Q&A session.
[Operator Instructions] With this, I would like to hand over to Mr. Gopal Vittal for the opening remarks.
Good afternoon, ladies and gentlemen. Thank you for joining this webinar to discuss Bharti Airtel's results for the quarter ended 30th June 2022. Also present with this -- with me on this webinar are Soumen Ray, Harjeet Kohli and Arpan Goyal.
I want to focus this quarter's earnings call on 3 things: some overall comments and an update on our performance; a deeper dive into our 5G approach now that the auction is behind us and why I believe we are so well positioned as Airtel to win the coming 5G game.
Despite macro challenges, we have seen another steady quarter from Airtel. Our consolidated revenues for the quarter grew sequentially by 4.1% to INR 32,805 crores. In India, our EBITDA margins improved to get to 51%. We've now received the $700 million from Google for a stake of 1.2%.
Our leverage at the end of the quarter stands at 2.52x and has been consistently getting better. This is due to our tight fiscal prudence and improved operating leverage that we are seeing. A quick word on ESG. To drive greater transparency around how our businesses are creating value by contributing towards a sustainable economy. We are voluntarily making our BRSI disclosure with effect from the financial year ending March 2022.
I'm also delighted that sustained analytics has upgraded the ESG rating of Airtel to low risk with a significant enhancement in rank across all telecommunication services companies globally. During the quarter, Nxtra was recognized as one of the sustainable organization in 2022. In order to bring world-class digital access to all and reduce the digital divide, we became the first operator to launch broadband or the first ISP to launch broadband in Ladakh and the Andaman and Nicobar Islands.
Finally, we continue to maintain the highest standards of corporate, financial and operational disclosure. Let me now briefly touch on each of our businesses. I'll start with our smallest but most exciting business. Airtel Payments Bank now has a monthly transacting user base of 44.4 million. This quarter, we clocked an annual revenue run rate of INR 1,100 crores and a GMV of $5.4 billion.
Our take rates of 0.66% are the highest in the industry, making us the only profitable fintech player. In addition, our digital services have now reached an annualized run rate of INR 850 crores. As I mentioned before, our digital business is extremely capital-light and leverages our underlying strengths.
Now I want to turn to Airtel Business, a segment that I have always called a jewel in our portfolio. Here we reported strong 4.4% sequential revenue growth. We have outperformed the other listed companies in this space and accelerated our market share further. In fact, Airtel Business has reached a special milestone and is now the #1 B2B player in India. Our success can be attributed to a razor sharp strategy, of both going wide to penetrate more accounts as well as going deep to serve our largest customers with a full set of products and solutions.
Clearly, the focus on fast-growing emerging products, including CPaaS, data centers and cybersecurity are now yielding results. I also want to call out IoT, where we are outperforming the market by a margin. You should also know that all our IoT customers are postpaid customers, but we include these connections as a part of the Airtel Business segment.
So at the end of quarter 1, we had a customer base of about 29.2 million on postpaid, 18.1 million reported in mobile services as a part of our mobile services and another 11.1 million of IoT connections. In effect, postpaid, therefore, as a segment is about 45% larger than our nearest competitor. Our broadband business, thus continue to see very strong customer additions driven by rapid rollouts and solid marketing.
During the quarter, we added 1.7 million home passes on the back of accelerated rollouts through the local cable operator model. With this, we are now present in 983 cities. We are making solid investments in this segment to build a pole position in the broadband market. As a result, we added 310,000 customers and reported 5.7% sequential revenue growth. This was despite some offset of a one-off benefit in quarter 4 full year -- in quarter 4, 2022.
The DTH business saw a decline of 0.9%, but the silver lining is that we have consistently grown market share over the last few quarters. As I've said before, the entire industry has been brought to its knees by excessive regulation. The new data order introduced a few years ago, created mind-boggling complexity for the customer with no benefit to any stakeholders.
Secondly, the same content, if provided through a different medium, broadband or wireless, is not subject to this tariff order creating an arbitrage and uneven playing field. Finally, good content, while it has come down, is still being made available free on free-to-air channels. This perfect storm has led to serious headwinds in an industry where there's still a massive opportunity to grow from cable.
Given this backdrop, our approach has been to leverage the core strengths of the ongoing broadband explosion to put Airtel Black and convergence at the heart of our strategy. We have seen substantial growth in the last month on our broadband and content bundles that includes linear TV as well as OTT content. We expect the business, therefore, to seize some recovery in the coming quarters.
Let me now turn to the Mobile segment. Here, we've seen sequential revenue growth of 3.4%. We added 4.5 million 4G net adds in the quarter and about 250,000 postpaid customers. With the increase in prices of entry-level smartphones and the focus of OEMs on higher-value smartphones, the overall industry did see a nearly 15% reduction in upgradation from 2G to 4G or from feature phones to smartphones.
We expect this to normalize over the coming quarters. At our 183 ARPU, we continue to lead in the industry. We are confident that we will see ARPU of 200 and eventually 300 arising out of tariff prices. In all the quarter witnessed broadband growth across all our businesses and our focus on warrant [ rates ] led to improvement in margins.
Let me now turn to 5G, and I do want to spend some time on this. As of now, Airtel has acquired 100 megahertz of 3.5 gigahertz spectrum across every circle in the country. 3.5 gigahertz, as you know, is the world class layer for 5G and will give us a pan-India footprint that can bring a true 5G experience for our customers. In addition, Airtel has acquired 800 megahertz of 26 gigahertz spectrum across every circle in the county.
This is a band that has limited propagation, but because of the massive spectrum tranche, gives you 4Gbps [ net ] speed. Finally, we are bolstered our spectrum holdings in the mid band and low band. The mid band, as we call it is the 1800 and 2100 band and the low band is a 900 band or the 850 band.
As a result, the total commitment in this auction was INR 43,040 crores. Let me now give you our approach on 5G. I want to do this in real laymen terms so that I'm hopefully able to demystify the incorrect narratives that are making rounds in a few places. You may know that there are 2 modes in which 5G operates. Standalone, or SA as it's called, is the mode -- is one of the modes and non-standalone, or NSA, is the second mode.
In the SA mode, 5G comes as a drop up to an existing 4G radio layer. The 4G layer then operates independently. There are two issues here: The first is a lack of a well-developed ecosystem for SA devices. The second issue is propagation. The workforce layer 3.5 gigahertz has lesser propagation than even 2.3 gigahertz, which is a TDD band, and this impacts coverage in urban areas.
As a result, SA can be effective only if there is a sub-GHz layer that is also offering SA and the two work in conjunction. The role of the sub-GHz layer is only for coverage. It is not for capacity or speed. So the sub-GHz basically goes and is needed where 3.5 GHz does not reach. In fact, we have seen in our trials that the sub-GHz layer on 5G SA give only 8 to 10 Mbps speeds. And therefore, it's no different from 4G.
This is a very important point to understand given the misconceptions about the 700 band and why it is a [ Pan-Asia ] for everything, it is not. The band is absolutely no different from 850 or 900 in terms of propagation. All it does is to provide coverage at the edge, deep indoor or at far flung areas and it gives you 4G-like speeds, nothing more.
The second mode that 5G operates is NSA. Now a quick word on the physics involved. The limitations of radios and spectrum propagation is always the uplink. It is the uplink that allows the user device to upload a photo or any other content or even have a 2-way voice conversation. The downlink, however, can travel much further, which, as you can imagine, is of no help for user experience because without uplink capabilities, there will be no coverage.
In the NSA mode, the great advantage is that the 3.5 gigahertz layer actually travels much further because it performs the downlink. It does not need to do the uplink because that is done by the 4G layer. This is invariably the 1800 or 2100 bands, also called the mid band. As a result, there are several advantages of the NSA model.
First, the 3.5 gigahertz band extends at least 30% more of the downlink, which implies 100 meters extra coverage. This gives you substantially more coverage in urban areas when mid-band is providing the uplink. At the edge, when you need further extension of coverage and the mid band doesn't reach, the fallback in terms of coverage is provided by the sub-GHz layer, 850 or 900 where we have a tranche of spectrum all across the country.
The second advantage of NSA is that all devices work on this mode. It is the most widely available ecosystem in the world. And as you know, telecom is a game of ecosystems. In the U.S. and South Korea, where both SA and NSA have been launched, the traffic on SA is less than 10% of total 5G traffic. The third advantage of this mode, it allows us to use existing 4G technology at no extra cost for the uplink or what's called the anchor band since we already have the radios and the spectrum that is live on our network.
Finally, the last advantage of NSA is around experience. It will allow for a faster call connect time on voice. In addition, it allows us to provide a faster uplink than anyone else, given our massive spectrum holdings in the mid band. And this is what leads me to the crux of the issue. To offer NSA, you need solid mid band spectrum, 1800 or 2100 because that's what provides great uplink experience.
Over the last few years, we have strategically accumulated the largest pool of mid band spectrum. Today, we have 30 megahertz of the mid band spectrum. This does not include the 2300 band TDD, 30 megahertz of mid band spectrum in 4 circles and 20 or above in all of the rest. Our competition does not have such large mid band spectrum. Do remember that if we did not have this larger chunk of precious mid band spectrum, we would have no choice but to buy the expensive 700 band spectrum.
And once we had bought it, we would have to deploy large power gasoline radios on this band. Not only would the costs have been higher, it would have led to more carbon emissions and very, very importantly, given us no additional coverage than our existing 900 spectrum band. So in sum, what would we have got? More cost, more ESG unfriendly and no improvement in coverage.
Therefore, this well thought out strategy for the spectrum acquisition through a combination of auctions, M&A and trading over the last 5 years has allowed us to avoid the need for adding an expensive sub-GHz band and yet deliver a world-class 5G experience. In sum, our consistent long-term spectrum strategy will help us meet all our objectives; the best 5G experience, 100x capacity enhancement, the most power efficient solution and the lowest total cost of ownership.
We believe this will give us an enduring competitive advantage for years to come. With this auction, we are fully confident that we will not be required to spend any material amount of spectrum for many years to come.
Let me now briefly touch on our 5G readiness and rollout. We intend to launch 5G starting immediately and extending to a pan-India rollout very soon. By March 2024, we believe we will be able to cover every single town -- 5,000 odd towns and key rural areas as well with 5G. In fact, detailed network rollout plans of 5,000 towns in India are completely in place. This will be one of the biggest rollouts in our history.
While our 3-year CapEx will remain around the same levels, this rapid rollout would see some advancing of CapEx on an EMEA basis. Every network domain is completely 5G ready. Our transport layer has been built over the years with a combination of fiber to the tower, synergies from fiber to the home and the availability of E-band microwave spectrum. Every site we roll out will be backhaul ready to provide 5G experience in line with what is needed to deliver a one-class experience. multi-Terabit MPLS and Internet backbone is fully ready to take on the 5G led data growth. Even on the cloud network side, we are best-of-breed [indiscernible] for the cloud platform that will help us in the deployment of our network in this cool manner. At the same time, our core and our radio network are on future-proof and can operate either on NSA or SA mode.
So over time, as more and more devices come into India, and we see disproportionate 4G traffic move on to the 5G network, we will be able to release all of our spectrum on 4G band, 1800, 2100, 900 and even 2300 to SA and switch seamlessly from one mode to the other, all this at the flick of a button through software. I now want to step back and share my confidence by giving you 5 powerful reasons on why we will win the 5G game.
First, our business is squarely focused on quality customers, whether it's mobile, broadband or converged products, everything we do across channels, marketing or differentiation is singularly centered on quality customers. We see the lowest level of churn among these customers, and they will be the ones who will upgrade to 5G device faster than anyone else. Second, we want to have one of the most powerful enterprise businesses where we have the relationships and enjoy the trust of our customers. Thus, business has incredible momentum. All of 5G enterprise innovation will be brought to the market by us to serve these customers. I believe we can win even more strongly on 5G because of the enterprise business.
Third, we have compelling digital capabilities now to lead the 5G game. Whether it's our omnichannel approach or indeed our digitization of network, we are now ahead. A couple of examples here. We now have our own Airtel self-optimizing network with [ fire ] engineers. This is an AI ML-driven in-house, closed-loop, self-healing platform. It was developed for automated network management for proactively managing forts or saving on the power on the fly and much else. Use cases are being written every fortnight. This powerful tool has won global awards as well. The second tool has been built to completely automate 5G network planning, configuration, onboarding and even field operations.
Our security center for enhanced security and vulnerability management is also 5G [indiscernible]. These capabilities will also allow us to deliver the best 5G experience. Fourth, we have a set of digital services, each of -- each of which plays a role in reducing churn and improving stickiness across both the consumer and the enterprise business. So whether it's our Payments Bank, our Wynk music platform, our Xstream video platform, our launch of Airtel IQ or even Airtel Ads. Each of this also creates stickiness. And this stickiness translates lower churn and higher switching costs. Finally, the most important reason, our track record. We came into the market late on 4G in 2016 against deep-pocketed competitor that rolled out the larger and wider network than Airtel, who was in the industry for many years. It had not happened anywhere else in the world. While that was happening, narratives were spun, but pure 4G network is so much better than the legacy network.
Simple and clean 4G is so much better than multi-technology networks. We have heard such illogical and factory stories before. All I can point to you is our track record. We have the best quality customer in industry. We deliver the best experience in the industry. We are constantly getting better every day. Our whole purpose is to serve our customers. We listen to them carefully. And the icing on the cake is that we have outperformed the industry by notching up consistent market share gains for many years.
In my view, the only thing that customers really care about is experience. On that front, we will deliver again and again, better than anyone else because we've been very astute in our spectrum acquisition because we will continue to invest in the best technology. We will work with the best global partners. We will not try and do everything ourselves because we are not arrogant to even entertain fleeting feeling that we can be better than our partners. These are the reasons for my confidence. And for you, I think the proof of the pudding will lie in the eating. With this, let me open up the floor for Q&A.
[Operator Instructions]
The first question comes from Mr. Sanjesh Jain from ICICI Securities. Mr. Jain, you may please unmute your side, introduce yourself and ask your question now.
Thanks for that elaborated understanding on 5G. That's really, really useful. A few questions on 5G. First on the CapEx, you did mention that your 3-year CapEx on a -- holistically will not change. So we assume that for 3 years, we still are confident that we'll be spending only INR 75,000 crores or INR 750 billion, which we were anticipating earlier. That remains true. How much of it will be front loading because we are also very, very aggressively pushing 5G, now talking of reaching 5,000 towns by end of March '24. That looks like an impressive plan. But how much will it consume in terms of cash flow in the near term, say, for next 2 years, how should we look at it? That's number one.
Number two, on the operating cost, you did mention on the user experience and the CapEx side of SA and NSA. Can you elaborate what will be the difference between the operating cost in SA and NSA, now considering that NSA will also [indiscernible] 4G asset while SA will be predominantly a standalone 5G stack. So will it have any material difference on the operating cost side of it? That's the second one. And third one, on the economics of 5G itself, now that we have bought such a large spectrum pool, and now we are also rolling out additional network on it. How should one think on this incremental investment? What will be the IRR or however how will an operator recover the investment? So these are my initial questions.
Thanks, Sanjesh. I think on the CapEx, we don't give formal guidance anymore. But if you look at our CapEx profile over the last couple of years, we expect, if you take a 3-year type of view, then we will probably be in the same ballpark. Of course, we will advance the CapEx in this year itself, starting now. So to that extent, there will be an advancement of CapEx that will happen over the course of this year. And I think it will sort of -- in an 18-month period, it will normalize. On the operating costs, I think if we had to -- and let me answer this question differently, if we did not have the mid band spectrum, and we had to buy the 700 band, spend the, let's say, 10 megahertz of spectrum, INR 40,000 crores. You need massive antennas.
You need a big radio to 3 extra radios across the country in all the sites that we put in, extra power that will need to be deployed there, extra rental for those radios. We think that we would have had a cost per GB disadvantage compared to where we are today, of almost 50%. So if you are at x today, we would have been 50% higher on cost per GB. That's our math, which is the reason that over the last 5 years, we have carefully acquired a large pool of mid band spectrum, which was very, very well thought through. We never went public about it because obviously, it was highly confidential, but this was really the strategy for us to make sure that we get to repeat 20 -- 30 megahertz in all the circles on mid band, and that is really what we've done.
The economics of 5G is a function of, I would say, the incremental tariff increases that the industry sees. Remember, over a period of time, 5G CapEx will be displaced. -- sorry, 4G CapEx will displace with 5G. Bulk of the rollout has happened. Today, already 8% to 9% of devices are 5G ready. Shipments that are coming in are about 30%, 35% of devices, and we expect this will change a lot. So by next year, almost 80%, 90% of devices will all be 5G devices that will come in, which means that 8%, 9%, if you project that forward to, let's say, March 2023, could be close to 13%, 14%.
And then by March 2024, it will be much higher. So a bulk of our rollouts are done which means that the CapEx that goes on to radio on 4G will get displaced by 5G. And this is why I say that over a 3-year period, our CapEx profile will fundamentally stay the same. How the pricing will work between 5G and 4G is something that we haven't yet quite decided. So I'm not going to comment on that right now. What we have seen globally is that 5G by itself is not yet giving incremental ARPU to any operator anywhere in the world. But in India, as you know, tariffs are still very low. We do expect tariffs to increase. And with every increase in tariff, obviously, the economics will change. The return on capital will get much, much better.
Yes, that's quite clear. Just one last question before I come back into queue. You have form a Customer Advisory Board for your enterprise business. Now what are they telling about the 5G adoption in India, particularly for Airtel? What are they demand? And are we in sync or our 5G CapEx is in sync with them? Are they really excited about 5G? Or they think it's too much for them to invest initially to justify their ROC? What is their understanding on enterprise side of 5G?
See on the enterprise side of 5G, I think the Advisory Board has given us a lot of ideas on new services to be launched. Airtel IQ came out of those discussions, the work-from-anywhere solution came out of those discussions. A lot of the work that we've done on SD-WAN, the acquisition that we made of a start-up that we acquired 25% stake is, again, coming out of those conversations. We are also having multiple conversations on private networks with a large number of companies. So we are looking at that. As of now, I think our rollout will happen and over a period of time, we will see how the Enterprise use cases play out, but there are ongoing conversations on standalone networks for large distributed enterprise.
So do we see in FY'24, some contribution in Enterprise coming from the 5G part of it? Or will it be an ambitious one?
No, no. I think that the Enterprise contribution has been going up to the overall profile of the company and the enterprise business has performed exceedingly well over the last decade. We see no reason why the Enterprise business will not sustain its growth. And obviously, as new technologies come, those will also get adoption. So to that extent, I don't look at isolating what is the Enterprise business coming out of 5G. I look at 5G, 4G and all of this as underlying connectivity layers, but also giving you the capability of adding new solutions. But ultimately, we want to deliver a solution for the customer, which is -- which will lead to revenue. So when -- whichever technology comes from, is not the issue. The issue is, are we able to grow the Enterprise.
The next question comes from Mr. Pranav Kshatriya from Edelweiss Securities.
My first question is on this NSA and SA. If you look at this SA rollout has been very recent. And to that extent, as you rightly pointed, the ecosystem is not mature. But you -- do you think that at some point, if the SA becomes more predominant and the ecosystem picks up, we have the flexibility to really acquire additional spectrum and make the network SA? That's my first question. Secondly, there has been a fairly strong acceleration in terms of number of cities where you are present in terms of the fiber for home broadband. So can we expect that to drive disproportionate growth in the coming quarters? And lastly, a bookkeeping question. Can you give some color on how much is the depreciation pertaining to the lease liabilities because there is a sharp increase in the depreciation in this quarter. So some color on the breakdown of that. Now these are my three questions.
So I want to explain the SA versus NSA issue, and let me just kind of repeat specifically for the question that you asked. Over a period of time, as more and more 5G devices come in, all 4G traffic will shift on to 5G networks. Spectrum that we have in the mid band, in the 1800 and 2100 band, which is 20 to 30 megahertz, will be the first band that will get reformed for 5G. 2300 band will also get reformed for 5G. Ultimately, even the sub-GHz band, which remember, only gives you coverage, nothing more. That coverage will give you -- if it's -- if you were 10 megahertz of spectrum, it will give you about 7 to 8 Mbps. If you were 5 to 6 megahertz of spectrum, it will give you 3 to 4 Mbps. So all it does is it gives you coverage. It's 4G-like speeds will be the last band to get reformed on to 5G. As the reforming happens, the NSA mode can seamlessly switch on to SA. Everything is ready. The core will be ready, the radios are ready.
They're all software enabled. So it's just a mode of operating, and in a way, there is absolutely no difference in what we will ultimately deliver in terms of customer experience. On fiber broadband, I think we are seeing a sharp growth. We've added 310,000 users this quarter. And we see that this growth will hopefully sustain. The demand for home broadband is exploding coupled with convergence and the propositions that we've launched on Airtel Black, where we are combining the content bundles both OTT and linear, we expect this to continue to see significant traction. And hopefully, this business, which is a very good business, with very low levels of churn will become a very sizable business in the years to come. I will leave the last question to Soumen to address.
Yes. Pranav, depreciation has gone up by about a couple of percentage points, of which there is a day extra in this quarter, plus the stock of number of towers is keeping on increasing. Some renewals has happened, but nothing out of the ordinary.
No, I just wanted to understand because you had this renewal, which would have had sort of a one-off impact because the duration increases and that disproportionately increases the depreciation for those lease liabilities. so if you break it off, it will be -- as well.
No. So the -- essentially, the impact is more on interest because, as you know, in equity monthly installment, initial period, the principal amount is much lesser and the interest amount is much more. So the impact is initially much more on interest and less on depreciation. But overall depreciation up by about 2 percentage points, 1% out of that is because of a number of days. The other is just routine [ procedures ].
The next question comes from Mr. Kunal Vora from BNP Paribas.
This is Kunal Vora from BNP Pariba. Question is on the home broadband opportunity. Can you talk about the fixed wireless broadband opportunity in India? What the potential size of the overall home broadband market and whether we can expect fixed wireless broadband to reduce the size of FTTH opportunity. Also, like home broadband, this quarter looks a little soft. I understand like you mentioned about the high base from last quarter, but any -- are you seeing any other issues there with the economy opening up? And are the customers down trading with more -- like work from home [indiscernible].
So on home broadband, I think that we have been adding about 300,000-plus users -- customers every quarter. So we are sustaining that trajectory. If you're commenting on the sequential revenue growth, then yes, as the base in [ place ], you start seeing some -- not a softening, but the sequential growth, not in absolute terms, it's about the same, but in percentage terms comes down. But we're still adding 300,000 users. And we, like I said, I have confidence that this trajectory will get sustained. I think fixed wireless is an interesting opportunity and has been deployed in some markets in the U.S., even Germany. But one of the things that we are seeing on fixed wireless is, ultimately, the cost for home pass. It has to be -- it has to sort of work competitively versus fiber in terms of cost for home pass. In India, if you look at our cost per home pass, we have been able to bring it down substantially.
In the cities where we roll out ourselves, the cost for home pass is down to about $30 to $35. In the LCO cities, it's a fraction of that cost. So on a blended basis, it's very much lower than $30 -- substantially lower than $30. Now typically, if you roll out 100 home passes, we see at the end of 18 to 24 months, about 30% utilization, which means that if you're taking a blended cost of, let's say, $15 to $18, then this is about $50 cost per home -- connected home. The cost of the router on fixed wireless today is about $200. In the case of Verizon in the U.S. or [ DT ], the cost per home pass for them is almost $400. So for them, the fixed wireless actually makes a lot of sense. You can put a router and spend that money.
So as the scale in India develops, we will need to see the cost per router for the fixed wireless actually crash below $50 then it will start making some sense. But until such time that, that happens, I think our objective will be to really go very, very aggressive on fiber rollout itself because there's no better substitute for home broadband when compared to fiber. Fiber is always going to be better than wireless because it's a dedicated pipe that you're delivering to the home and you have full control over both the uplink as well as the downlink with no constraints on traffic or population in that particular area. So I think it's down to the economics as the way we see.
Understood. Second question is on the SG&A cost. We've seen a fairly sizable increase, 25% year-on-year, almost 50% over the last 2 years. And the churn has remained elevated over the last few quarters. It went to like 2% levels, but back to 3%. Can you talk about like what's happening on the competitive intensity? And when would you expect this to moderate?
Yes. Maybe Soumen, you can cover this.
Yes. So the SG&A increase is in line with growth in business and the competitive activity in the market. Also, there is a bit of deferment impact in this. It is not very different because when we started doing the deferment, now that cycle is getting catch -- caught up. But if you look at purely sales and marketing expenses, it has remained almost flat.
Okay. Okay. Okay. And lastly...
No, just to -- just to finish that discussion. If you are looking at other expenses inclusive, then there, there is an increase of close to about INR 400 crores, which is essentially around three items: The first being a -- there is this FLO liability recasting, which has been about INR 200 crores. And then Africa, there has been an increase of over INR 150 crores. Intrinsic India business, there has been a marginal increase of about INR 50 crores. So that explains.
Okay. Okay. Understood. And just one last question. If I look at the growth differential between Airtel and the #3 player, you've seen some narrowing of the gap. Airtel used to grow much faster compared to the #3 player. We've seen that narrow despite larger investments in superior network. Can you share your thoughts on why this is happening? And with larger 5G investment, would you expect accelerated market share gains from the #3 player?
I think that I don't want to comment on the #3 player, but I do believe that any operator that actually goes out and roles a 5G network, you will start seeing the best quality customers first sort of move to those networks. And we are very confident that our postpaid business, our converged strategy, high-value prepaid, all of this should see substantial growth with the rollout of 5G.
The next question comes from Mr. Manish Adukia from Goldman Sachs.
Looks like Manish Adukia has dropped off. We will move to Vivekanand Subbaraman from AMBIT Capital.
Yes. This is Vivekanand Subbaraman from AMBIT. A couple of questions. So first one is the circles where you have some administrative spectrum coming up for renewal in 2024, like UP, East Bengal and others, do you envisage spending meaningful amounts of money in the next 3 years in subsequent spectrum auctions? So that's question number one. Question number two is on the B2B growth acceleration. So Gopal, if you could just touch upon these factors that are responsible for the faster growth. I'm referring to the year-on-year growth, which has been on an upward trajectory and this is not just the current quarter, obviously. We've seen this play out in the last several quarters. So perhaps if you could help us understand whether it is because of the market growth itself, market share gains? And if you can just give us a bit more color or an update on the On The digital products and the new products that you launched in the B2B side and where the revenue contribution is coming, from those areas?
As I mentioned, that we believe that there's very, very little amount of spectrum spend that will happen over the many -- next few years. There are a few circles. There are very small number of circles now, some are the ones that are coming up for renewal in 2024, which still has some adding spectrum, but we have already bolstered a substantial amount of spectrum, even in the low band, Like, for example, if you take Odisha, we bought liberalized spectrum even though we have admin spectrum. We did the same in Bengal. We've done the same in UP East, where 6.2 megahertz of 500 will come up for renewal, but we already have [ 0.5 ] megahertz of liberalized spectrum that we bought. So we are trying to bolster our holding the same thing we did in Northeast [indiscernible].
So I am quite relaxed actually about the renewals component in some cases -- in some places wherever there is some traffic or requirement or so on, we may bolster it, but it is not essential since we are sort of running it on existing liberalized spectrum already. B2B has been a carefully crafted strategy as far as we are concerned. And let me kind of comment in two parts: Number one is that we have really retooled entire go-to-market on B2B. As mentioned before that 80% of our revenues comes from 20% of our accounts. These are the most important accounts for us, but I see as a glass half full. So the 80% of customer -- revenue that comes from [indiscernible] accounts, we are working with them through the Customer Advisory Board to bring more solutions, more services.
This is where Airtel IQ, the CPaaS [indiscernible], cloud -- sorry, cybersecurity, which is Airtel SECURE, SD-WAN and so on and so forth are all coming in to actually create a greater share of wallet. We've seen some fabulous stories where we were able to raise our average revenue per account quite substantially through concerted effort. The second part that we have is the 20% of revenue, which comes from 80% of accounts. This is where we have, again, retooled our GTM to really have a bunch of people looking at hunting for these accounts, and this is now true for the large enterprises as well as the top end of the medium enterprises, and I'll come to that in a moment. So I think that part, we are tracking very closely. How many accounts have we added? What is the average revenue per account for each account? So that's the second part of our strategy.
Third part is really around the SME side, where we spend a lot of money and time to build a digital marketing tool to create leads. So we are now getting almost 30% to 35% of leads coming from online and digital, which then goes to actually drive the SME growth, which is the third part of our approach. And the fourth part of our approach is that we have in-sourced or we have done away with the channel. The SME business is to be run by the channel and the channel would get commissions. We have now done away with the channel and we have about 1,200 people who we have brought into our own company, who are responsible for the SME side. So I think that entire go-to-market architecture is something that has been a very important reason for our growth. And while we've done this, we have taken out one layer in the organization so that the customer -- people who are serving customers are closer to the leadership.
The second part on B2B, which is a second block of act, this is a GTM part, it's the go-to-market. Second part is really to develop more and more solutions and services, and this is where we have spent a substantial amount of our time through building our own services as well as through partnering, taking small bets, acquiring a few companies to create a stronger portfolio of services. which can also build our B2B business. So Airtel IQ is one, Airtel SECURE is another. Airtel IQ has been building strong traction. Our CPaaS business is looking very strong. So I think it's the whole combination of all of this on the B2B side.
Very helpful. Just one small follow-up as far as the CapEx and spectrum purchases are concerned, we have not called the rights. The second call hasn't yet happened despite participation in the 5G auctions, so I'm just trying to understand what are the instances in the next few -- next 12 to 24 months, where we would call that. What would that be used for now?
Soumen, you want to take that or Harjeet?
Yes. Vivek, this is Harjeet. I think big rights, about INR 15,000-plus crores of residual calls are pending. And clearly, in the coming few months, we need to evaluate basis the business cash flow profile, which, by the way, now is organically both significantly positive in India and Africa. At what point in time should we call for it. Gopal also mentioned the ongoing dynamic play of some bit of acceleration of CapEx. So we will stay agile to what this business cash flow profile is and accordingly take a call in the coming few months on the residual calls for rights.
The next question comes from Mr. [ Abhiram Iyer ] from Deutsche Bank.
This is [ Abhiram Iyer ] from Deutsche Bank. Congratulations on good set of results. Sorry, my query was pertaining to the -- you mentioned that you're going to bring forward some of the CapEx cycle over the next 3 years to maybe the next 12 to 24 months due to the rollout of 5G. May I know whether this could change your debt profile and your leverage targets? Do you expect to end the year maybe sitting at a higher leverage than what it is currently? And the other thing is, obviously, there is a significant amount of operational cash flow available to the company given the nature and the quantum of CapEx that might be required, do you envisage coming to the debt markets?
Firstly, let me just kind of say that the operating cash flows of the business are strong enough for us to fund any requirement of CapEx, whatever advancing we need to do. Yes, the debt on paper will get larger because of this large spectrum payment that will be due to the government, but the way we look at spectrum payments is that this is a payment annual, which is over 20 years. So it is no different from a operating cost and ultimately, all the spectrums translate to revenue. So if you were to take that operating cost of the repayments that are due out of the INR 43,000 crores, I don't recall the exact number, but whatever that number is every year to [indiscernible] sorry.
INR 3,600 crores.
INR 3,600 crores, and you restate that, and you said that INR 3,600 crores is really to be paid every year and the INR 43,000 crores doesn't go into debt, you pull the EBITDA down by INR 3,600 crores, which is really like being served out of the existing EBITDA pool, then the debt profile is actually getting very, very healthy, given the operating cash flows. So the leverage position is getting better and better every year and has been so over the last couple of years.
Got it. And just a follow-up question more in terms of bookkeeping. But -- so given the fact that you called some of your foreign exchange bonds, is this going to be strategy seen going forward given that the FX rate has not been favorable for Indian companies? Can we see you looking more to call some of the foreign debt and try and get more onshore debt?
I'll just quickly pitch in. And just linking back, Abhiram, to what Gopal mentioned in the last question, if you see the EBITDA less CapEx, at least till now, keeping aside some bit of acceleration globally, we will have about [indiscernible] of operating free cash flow, and our interest costs are well within $2 billion. So that's the -- and taxes are still fairly minimal. So we have significant free cash flow profile that allows us to keep the leverage reasonably well in check as also reducing fundamentally. Within that leverage, our FX leverage is significantly low. As you can clearly see, bulk of the Indian debt is DoT liabilities, the past AGR liabilities. There is a significant amount of FLO liabilities and only a smaller amount is really external debt, of which a much smaller amount is FX.
So in general, the FX is fairly well managed in terms of the percentage of the outstandings that are in foreign exchange. The foreign exchange bonds that we've been calling off, a, till date have been in Africa, which I think is very welcome. There are two, three reasons: One, of course, Africa is creating the cash flow profiles that are able to keep calling the bonds back. Second, as Africa is a listed company, stands on its own. Increasingly, the guaranteed bonds from Airtel India are going down. If you would go back about 3, 4 years ago, maybe prior to listing up Airtel Africa, we had more than $11 billion to $12 billion of guarantees given from Airtel India for the bonds of $5 billion to $6 billion that were outstanding in Airtel Africa.
Now that has gone down significantly. So that's another reason why that buyback is important. Third is we have to time the interest rate situations in the market, which essentially means when the bonds can be bought back at a reasonable savings -- economic savings for the company is, I think, the right time. So we would not go just to reduce the FX leverage. I think the economics of the situation, the refinance costs and including the cash flow generation from both the engines, Africa and India, will govern the further buyback actions. But in general, the company maintains capital markets access, and we -- at least in the rupee market have been fairly active. And if required, we'll be active in the dollar markets too.
The next question comes from Mr. Aliasgar Shakir from Motilal Oswal.
This is Aliasgar from Motilal Oswal. A couple of questions. Gopal, first one on 5G. So you did mention that the ecosystem is yet to develop on the handset device side, we would probably, in the next couple of years still be somewhere about mid-teens kind of penetration levels. Just trying to understand what is the need to accelerate 5G rollout? I mean, is it to do with kind of matching competition? Or you think that, I mean, with our experience in 4G, we don't want to kind of stay behind the curve. Or do you think that there is some possibility of better monetization opportunity left? So I'm just trying to think about -- or why can't we kind of wait for some time led the ecosystem develop and then kind of accelerate.
We already demonstrated that in our 4G rollout where we did accelerated quite significantly well. And second question is, I mean, you mentioned that we have not thought about how we will monetize 5G, but I mean it's been 6 months since the last tariff, now that we plan to do 5,000 town rollout. How should we think of your price tariff hikes? Should it be direct tariff hikes or should we see change in the plans from these unlimited plans that we have to more kind of limited plans or 5G related kind of pricing? Yes, those are my two questions.
Well, I think that on the tariffs, as I said, tariffs need to go up, when it will go up and so on and so forth, I think I can't comment on right now. Secondly, to change the construct of plans -- price plans industry on -- for us on a unilateral basis is always a challenge competitively. And so we have to be competitive in the market. Ultimately, this business -- if you grow market share, you are in actually a much better long-term position for investors as well as for customers and employees. So to that extent, we will do the right thing to remain competitive. I think we have debated this. We can finnesse this a lot. We can sort of wait for the ecosystem. Ultimately, it's just pulling the same CapEx that anyway would have gone. I mean nothing is going wasted. It's just putting it there a little ahead of time so that we are present where we need to be.
Of course, having said that, there's a lot of analysis that happens on where we should go and which sites we should go to and all that based on devices and all of that. So all that is happening. But we did see, for example, in some of the C categories of those -- when we were a little late on 4G, we lost a little bit of share early on, and then we've now recovered that market share over a period of time. But in the cities and the A category towns where we launched head-to-head with our competition, we had a much better outcome. So I would say that it's just a question of bringing forward CapEx without trying to finnesse this too much and intellectualize this too much.
I think that's the way we see. You're right. If this were a market that was uncompetitive and it was a benign kind of market with very little competitive intensity, you may have said we can even buy the spectrum next year, because there's nothing -- it's not going away. I mean -- so to that extent, I think the fact is that we believe that we need to be leading the narrative in terms of what we do with our customers, make sure that our best quality customers not just stay with us, but we're able to attract good quality customers on to our portfolio. I think that's really what we're trying.
Due to time constraints, I would now hand over the proceedings to Mr. Gopal Vittal for closing remarks.
Well, I do want to thank you for tuning in. I think I did want to spend a substantial amount of time on 5G and give you the confidence as to why we believe we are very, very well positioned. I hope I've been able to do that in the last hour, and I hope also that we've been able to clarify to you what our overall approach is on 5G and why some of the narratives that are doing the rounds are probably not based on fact or not founded on fact. And I hope that's given you the confidence. So with that, let me sort of sign off and see you next quarter.
Thank you, everyone, for joining us today. The recording of this webinar will also be available on our website for your reference.