Wednesday, March 10, 2010

Taking On Mr. Market!

I have bought 2,000 shares of Bharti Airtel on 2 Mar 2010 at Rs. 277 per share and plan to buy a significantly higher quantity if the stock is available at similar or lower prices. I think this stock has significant upside potential with low risk. My logic for this purchase is laid out below. (Its a bit long - but then i plan to commit a significant amount of my net worth in this company)


TABLE 1 (click to enlarge - recommend opening in new window)

The table above captures the Revenues, EBITDA (operating cash flow), PAT and Market Capitalization of Bharti over the last 7 years. I have forecasted the 2009-10 numbers by simply doubling the half yearly revenue as of Sep’09 (which is in-line with its Q3 numbers). An analysis of this table reveals a lot about the telecom industry in general and Bharti in particular.

Bharti experienced tremendous growth in the last 7 years. Its revenues went from Rs. 50 billion ($1.1 bn) in 2003-04 to Rs. 396 billion ($8.7 bn) in 2009-10. That’s an 8x jump!

Similarly EBITDA and PAT went from Rs. 17 bn ($0.37 bn) and Rs. 5.8 bn ($0.13 bn) to Rs. 166 bn ($3.61 bn) and Rs. 97 bn ($2.11 bn) respectively in the same 7 year period. Again, that’s a 10x jump in EBITDA and a 16x jump in profits.

In the last 7 years Bharti generated about Rs. 597 bn ($13.0 bn) of cash from operations and Rs. 318 bn ($7.0 bn) in profits (see blue boxes). However, none of this cash reached the shareholder in the form of dividends. The main reason for this was that Bharti ploughed back all of the $13.0 bn to build infrastructure – towers, underground cables, generators, sub-stations etc. as it increased it’s reach to the smaller towns and villages of India. No other telecom company (barring probably BSNL) has spent so much on building ground-up telecom infrastructure – no wonder Bharti is the largest owner of telecom towers today.

So, how was the shareholder benefited by all this growth if no cash ever made it to his hands? In the first 4 years of the analysis, the market cap of Bharti went from Rs. 256 bn to Rs. 1,665 bn (See brown box) which means that if someone had bought Bharti on 1st April 2004, he would have made 6.5x his capital on 1st April 2008 – that’s a 550% return in 4 years! By not giving money to shareholders and reinvesting it in the business, the company and management actually made the shareholders much richer.

However, all this was the past. All good things come to an end and the same seems to be true for the Indian telecom industry. At ~500 mm users, most urban cities have crossed the 100% penetration limit. Most of the new growth is coming from rural India (as per Bharti’s latest annual report 60% of all new customers are from rural India) which does not spend as much as urban India. In the last 2 quarters, Revenues and EBITDA have been flat for Bharti – which means new customers are not adding enough to the revenues and old customers are spending lesser.

If lower growth were not enough, the industry is now plagued with competition. This is generally the case when the going is good and there’s money to be made. Till Jan 2008, there used to be only 3-4 players in every circle. In Jan 2008 the government issued new licenses and spectrum to 6 new players – Unitech-Telenor, Shyam-Sistema, S-Tel, Swan-Etisalat, Datacom and Loop Telecom. The government also issued new licenses to existing players like Idea, Tata and Spice in circles they were not present before. Today, there are 12 telecom operators in the country of which 10 have a pan-India license which means that potentially there could be 10 service providers to choose from in every city.

This is going to put tremendous pricing pressure on incumbents since new players will try and garner as many of the new subscribers as possible and try and convert subscribers from other service providers using lower prices. Tata Docomo, Uninor and MTS have already started the price war by coming out with a 30p per minute scheme and forcing others to follow suit. This will impact both revenues and profits of Bharti and other incumbents. All this has been telling on the company’s stock market performance. The chart below shows how Bharti has continuously under performed the market since Jan 2009. While the Sensex has gained 73% to 173 Bharti has lost 19% to 81.

In the short term, the merger with Zain Telecom has also spooked the market, and since announcement in early Feb the stock has tanked 15% from Rs. 320 to Rs. 275. Markets generally tend to react negatively to big cross border mergers as seen in the Tata – Corus, Tata – Jaguar deals. However, all these stocks are now back to their old prices.

Bharti currently trades near its all-time low of Rs. 280 per share with a market cap of Rs 1,100 bn. It’s P/E ratio is at ~11.0x as Mr. Market has come to the conclusion that growth in the stock is more or less over. This is a classic case of a big corporation being “currently out of favor” with the investment community. Ben Graham and Buffet have taught me that these situations present interesting long term investment opportunities and generally merit a thorough analysis of the company and its operations to see if the market is mispricing (as it frequently does) long term value in the current stock price. In the 1937 edition of Security Analysis, Ben Graham wrote:

“It is customary to refer with great respect to the 'bloodless verdict of the market place', as though it represented invariably the composite judgment of countless shrewd, informed and calculating minds. Very frequently, however, these appraisals are based on mob psychology, on faulty reasoning, and on the most superficial examination of inadequate information”

In our analysis, we need to understand 3 things:

1. Are the long term economics of this industry intact?

2. Is Bharti positioned to take advantage of these economics and does it have the competitive advantages that make it better positioned as compared to its peers?

3. What is the right price to pay for the inherent long term value of Bharti?


If the last few quarters are any indication, revenues and profits from mobile telephony have more or less plateaued. The growth story here is, for all practical purposes over. If we see the same table on top, the red circled CAGRs on the right (cumulative average growth rates) tell the story. In the last 6 years Bharti’s wireless revenues grew at a CAGR of 48%. In the last 3, 2 and 1 year the rates were 32%, 22% and 8% respectively. The only growth here can come from rural India which is going to be nowhere as exciting or sexy as that experienced by urban India in the last decade.

As investors, we can assume that mobile (or wireless) revenues are going to remain flat or show very little growth in the next 5-10 years. For Bharti, this is going to be a cash cow spitting out Rs. 100 bn ($2.2 bn) of cash EBITDA every year (since incremental capex on towers is going to be much lower). Bharti knows this as much as anybody else. It has to find newer un-penetrated markets to deploy this vast cash pile or it will remain a cash machine doling out huge dividends to its shareholders. Thus, the rush to enter new markets in Africa, Bangladesh and Sri Lanka.

The other indicator of future growth is the rush of foreign operators to enter India. As a Telenor or a Docomo or for that matter Etisalat, why would I want to enter a market that is saturated unless I saw value there? All these firms must have done their own research to see the potential for growth in India. However, my own experience of seeing foreign PE funds investing in India has made me fairly cynical about the quality and depth of their research. The other day I read an interview of the Sistema CEO saying there is a 700 million untapped subscriber base in India. I guess he was talking about India’s 1.2 bn population minus the 500 mn subscribers we currently have. Taking decisions on the basis of that number is a bit naïve as it assumes the whole of India’s population will start using cell phones in the future. Included in that are people below poverty line, children and senior citizens.

I think the real growth, if any should come from non-voice services because this is one area where there is still scope for growth. Around the world, telecom companies seem to derive almost 25% of their revenues from non-voice services like net browsing, data transfers, wimax etc (see the 2 graphs below by Motila Oswal and Deutsche Bank) whereas only 10% of revenues in India currently come from these services. Bharti’s non-voice revenues stand at 9%. With the upcoming auction of 3G and WiMax airwaves this segment is set to ride the next growth wave. Bharti stands to gain here more than others because of its captive subscriber base of 110 million users and lower incremental capex on infrastructure.

Source: Motilal Oswal Research (Jun 2009)

Source: Deutsche Bank Research (Feb 2009)

Passive infrastructure is the other area of growth for Bharti. Being an incumbent, it enjoys superior infrastructure which it can share with new players who cannot spend billions on infrastructure as it did. This leads to infrastructure sharing which will bolster the revenues of Bharti. Having spent huge capex over the last 10 years it can sit pretty as new incumbents lease tower space from it and pay rents.

I think these are the core areas of growth which can lead to higher revenues and profits in the next 5-10 year period.

This answers our first question. Though the next 5-10 years is not going to see the kind of growth the last decade witnessed, there is still some milk left in the udder. Maybe it’s going to be 10% or 15% and not the 41% of the last 7 years but its definitely there.


In Buffet terminology this is the “moat” that keeps away potential competitors from coming in and destroying shareholder value by eroding margins and growth. This is the key to long term value investing as it helps protect investors like ourselves from nasty surprises in the future. I want to spend some time here because the next few points are going to help me sleep well when the market goes crazy (as it will) in the future.

Competitive Advantages / Barriers to Entry


Being the earliest entrants, Bharti, Vodafone and IDEA have been allotted spectrum in the 900 Mhz band. Among the three, Bharti has the highest (15 of the 23 circles) allocation in the 900 Mhz band followed by Vodafone (9 of the 23 circles) and Idea (9 of the 23 circles) while none of the new entrants – Tata Docomo, Unitech Telenor, Swan and BPL have 900 Mhz spectrum. All new entrants have been given spectrum in the 1,800 Mhz band.

A little read on the internet and broker reports tell us that the 1,800 Mhz band is commercially less viable as it requires double the number of towers for similar coverage. This is because network signals in the 1,800 Mhz have half the wavelength Vs 900 Mhz. Thus, incremental capex and opex is going to be double here for new entrants leading to massive capital expenditure and lower margins. This a double whammy for new players as it affects both the balance sheet as well as the income statement. Motilal Oswal estimates the impact of this to be 10-15% on the bottom line. Also, new entrants would have to price their product at a much lower price (what Tata Docomo, MTS and Uninor are doing now) in order to get higher subscribers. This will put even more pressure on their bottom lines till eventually they will have to increase rates or go out of business.


I think there are 4 aspects to this point as outlined below:

1. The Reach Aspect

As per their latest annual report Bharti Infratel (Bharti’s tower subsidiary) exclusively owns 27,548 towers in 11 circles and 35,066 towers in the remaining 12 circles through Indus Towers (where it has a 42% stake). Indus Towers is a JV between Bharti, Vodafone and Idea and owns towers in 16 circles (4 circles common with Infratel, 12 circles on exclusive basis). Therefore, together with Indus, Bharti owns towers in all 23 circles in India. This is the highest across all private telecom operators in India.

Bharti covers 5,060 census towns and 414,906 non census towns and villages in India covering 81% of the country’s population. Its national long distance infrastructure consists of 101,337 route kilometers of optical fibre while its international infrastructure includes ownership of the i2i submarine cable systems connecting Chennai to Singapore and joint ownership of transatlantic and transpacific routes. Any player, especially new (and on the 1,800 Mhz band) will find it extremely difficult and costly to replicate this network reach (up to 5 years or more as per industry analysts). Higher network coverage also means quality of service (roaming, network and data coverage) are far better for a customer using Airtel compared to a new entrant or an incumbent with lower reach.

2. The Value Aspect

In Feb 2008 KKR bought a 2% stake in Bharti Infratel for $250 mm implying a valuation of $12.5 billion for the whole company (see table below). However, Feb ’08 was the peak of the market and valuations were most probably stretched. Having said that, as late as Dec 2009 Citigroup sold its stake to JP Morgan at cost for $50 mm valuing Bharti Infratel at $10 billion. If we go ahead and buy Bharti shares we will automatically become owners of the tower company as well. Therefore, while analyzing Bharti, we will treat the Infratel arm separately by subtracting its “value” from the enterprise value of Bharti or by treating it as an ‘extra gift’ that we get with every one share of Bharti.

Table 2

3. The Cost Aspect

The setting up of a tower involves both passive (tower itself, repeaters, shelters, generators etc) and active infrastructure (electronic components on the tower such as antennas, feeder cables, nodes, radio access networks etc) spending. As per industry experts, the cost of setting up one tower is Rs. 30 lakhs. Clearly, new entrants will have to spend substantial amounts of capex on the construction of these towers. Amongst the incumbent private players, Bharti has spent the most on the setting up of these towers across the country. As we pointed out earlier, Bharti has spent close to Rs. 596 billion ($13 bn) in developing this infrastructure in the last 7 years alone.

4. The Revenue Aspect

As a new player it makes more sense to lease towers from incumbents than to spend on constructing them. This does 2 things – 1) makes new players dependent on incumbent players and 2) generates additional revenue for incumbent players. Also, new players still need to spend on active infrastructure to cater to their customers, so it’s not exactly ‘capex light’

For the year ended Mar 2009 Bharti’s revenues and EBITDA from passive infrastructure (tower business) were Rs. 42,489 mm (9% of total revenues) and Rs. 15,022 mm (9% of total EBITDA) respectively, a growth of more than 700% from Mar 2008. As new players enter the market, revenues from this segment are bound to go up even higher.


At 141 billion minutes, Bharti was 5th in the world in the number of mobile minutes carried on its network for the Apr-Jun 2009 quarter (see table below) and almost double its nearest competitor in India - Vodafone. This is a significant achievement given the fact that compared to the other players; Bharti is a single country operator (Vodafone operates in 30 countries). Despite such high volumes, Bharti’s EBITDA margin from the mobile phone business in FY09 was 41%. This gives you some indication of the scale and cost structure that Bharti has built-up over the years. As per Macquarie, it is “likely the lowest cost producer of voice minutes in the world and this remains the biggest entry barrier for greenfield players”. This is a huge benefit for the investor as this provides a safety float around the company which makes it difficult for competitors to match its products and services at the same cost.

On 14 Dec, 2009 Economic Times ran an interview of Vodafone CEO Vittorio Colao. Let me quote verbatim a few lines from the interview that will give you an idea of Bharti’s scale advantage.

Q: In your last quarterly results, you have said that despite the price wars, Vodafone will be looking to leverage its brand and scale in India. How do you plan to do so? I believe your operating costs in India are much higher that that of your closest competitor Bharti Airtel.

A: …In terms of networks, we have been able to use our global scale to get the best deals from western and eastern equipment companies. Yes, Bharti has scale and they are ahead of us, but we are patient. Give us another 15 YEARS and we will catch up with them

Q. You must be joking about the 15-year period.

A: I am serious. In our sector, it takes time. I am not obsessed with being number 1, but we are happy to be among the top two leaders of the market here. Bharti is a very well run company. I admire them. But right now, they are well ahead of us

These words coming from its biggest competitor are manna for the investor’s ear!

Source: Macquarie Research (Oct 2009)


Airtel was voted the 2nd most trusted brand in an economic times brand equity survey and 8th in the Top 50 most valuable brands in the world (by Brand Finance Plc). Over the years, Bharti has spent substantial amounts of money on marketing and advertising the ‘Airtel’ brand resulting in significant brand recall amongst consumers. In FY09 alone it spent $521 mm (Rs. 2,500 crore) on sales and marketing.

New players will find it extremely difficult to match the advertising budget of large scale players like Bharti and Vodafone who have already built up a significant brand presence throughout the country. Also, as an investor in Bharti it’s heartening to know that it is this brand that makes ordinary subscribers go to local kirana stores or airtel outlets month after month to recharge the currencies in their cell phones.


Sunil Bharti Mittal is the founder, owner, chairman and managing director of Bharti and owns close to 45% of the company. A short wikipedia search reveals that he’s a first generation entrepreneur who started his first venture as early as 1976 with a capital of Rs. 20,000 making crankshafts for bicycle manufacturers. After selling his bicycle parts business in 1980 he moved to Mumbai and stared a generator importing business and in 1984 started assembling mobile phones in India. He got his first big break in 1992 when the government was auctioning mobile phone licenses in Delhi. Understanding the potential of mobile connectivity, he bid and won 1 of the 4 auctioned licenses and has since built Bharti into the largest telecom operator in India with more than 110 million customers and $8 billion in revenues.

No doubt, Sunil Mittal has the vision and the entrepreneurial flair required for success but the best thing about him is that he has built-up and maintained a squeaky clean image of himself and his company. In the post Raju era, an investor needs to be doubly sure that the person at the helm of affairs is passionate and committed (financially and emotionally) to creating shareholder value. He is also conservative (Bharti has very little debt with most capex funded by internal accruals). All this has led institutional investors to own close to 27% of the company.

One of the best tangible ways to measure management performance is to see how effectively management reinvested earnings to create shareholder value. Buffet always says that “for every dollar retained by the corporation, at least one dollar of market value should be created for owners”. I did this exercise for Bharti for the last 7 years. The results are laid out in the table below:

Table 3

In the last 7 years Bharti retained Rs. 318 bn from shareholders and reinvested it into the business. At the end of 7 years, Bharti’s market value increased by Rs. 788 bn. That’s Rs. 2.5 of value for every Re. 1 invested in the business. The other option would have been to return everything to shareholders so that they can invest that money on their own. Clearly, in this case it would have been far more profitable for shareholders to invest it back in the company than to invest it on their own. This number gives you confidence in the management especially in light of the recent Zain merger. If Bharti can reinvest future earnings and create similar value, it could mean a windfall for equity shareholders in the future.

These 5 competitive advantages create a fairly secure moat around the company. As shareholders I think, we can rest assured that the company is far ahead of its competitors especially with regards to its first mover advantage in spectrum allocation, the depth and reach of its network, the scale of its operations, its brand and finally its strong management team. These attributes should keep the company ahead of the pack for a few years to come. (15 years as per its nearest competitor).


Now that we know what the potential of the industry is and the inherent competitive advantages that Bharti enjoys, we need to answer the all important question – What is the right price to buy this value?

In the first table we saw how Bharti has performed in the last 7 years. Now let’s try and extrapolate this table 5 years forward and see, with the help of some basic assumptions, what we can expect:

We will make the following basic assumptions in our forecasts for the 5 year period:

1. The growth rate of profits, EBITDA and Revenue
2. The P/E ratio at which the company will trade 5 years from now
3. The value and timing of sale of the Tower business

Let’s start by assuming a 10% annual growth in profits over the next 5 years (see red circle). The 2009-10 profits is most likely going to be around Rs 97 bn. If we grow this by 10% every year it becomes 156 bn in 2014-15. Remember this number was 60%, 34%, 23% and 23.1% for the last 6, 3, 2 and 1 year respectively but of a much lower base. The 10% number may seem to be conservative given the fact that Bharti is going to earn incremental revenues from non voice services, Zain merger and tower revenues. There is also a strong likelihood that the industry will consolidate in the next 5-10 years with only the strong players remaining profitable.

However, we need to be mindful of the competition and risks involved. Here, it’s best to apply Ben Graham’s “Margin of Safety” approach which advocates that investors will never know everything about a given company. However much we try to analyse the risks, we will end up making errors on macroeconomic events, interest rates, regulatory changes etc. Like Buffet said "It is better to be approximately right than precisely wrong." We should err on the conservative side. 10% is not an unreasonable number given our discussion of future growth.

Our second assumption is that Bharti will trade at a P/E of 10.0x after 5 years. This seems reasonable given the fact that Bharti is likely to be a stable cash flow generating company after 5 years. A P/E of 10.0x implies little or no growth in the business going forward. I think this number too can surprise us only on the higher side.

Our third assumption is that the tower business trades at $9 bn, a 25% discount to peak valuation of $12 bn and a 10% discount to the most recent transaction, which was when Citigroup sold its Bharti Infratel stake to JP Morgan for an implied valuation of $10 bn in Dec 2009. We also assume that this is the same value at which this company trades after 5 years. Here again, we are being conservative. If the tower business is IPOed at a higher valuation this number could increase by 25 to 50%. However, our objective is to see if it makes sense to buy this company at the most punitive of assumptions. The tower valuation at $9 bn also implies that our going-in P/E is 7.17x (green circle) for the wireless business.

Armed with these 3 assumptions we can calculate the going-in price at which we make a reasonable return on our investment. Starting with Rs. 275, the model tells us that we can make a 12.4% IRR (78% return over 5 years) or 1.78x our investment in 5 years if these 3 assumptions turn out to be true. The stock price at the end of 5 years would be Rs. 481.9 (see green circle). A small sensitivity table below tells us what will happen at different growth scenarios:

If the growth in profits turns out to be 20% instead of 10%, we can make a 20.6% IRR (153% return over 5 years) per annum and 2.53x our initial investment (see the 20% column). For this to happen, the stock price after 5 years needs to be Rs. 685.

What I like about this sensitivity is that even if the company grows by only 5%, we will not lose money. We still end up with a 8.5% annual return. However, a closer scrutiny reveals that at 5% growth the retained earnings (Rs. 533 bn) will be much higher than the difference in market cap (Rs 491 bn) implying that Bharti will not be able to create even Re 1 of value for every Re 1 invested. We can, at the least, expect Mr Mittal to create value equal to reinvested earnings. Otherwise he is better off declaring all profits as dividends. From our earlier analysis (Table 3) we also know for a fact that in the last 7 years he reinvested Rs. 318 bn and returned 787 bn or 2.5x. Even our conservative 10% growth case assumes that he will only create Rs. 785 bn of value for Rs 617 bn of investment or 1.3x (see green circle).

Now, if we do the whole analysis at Rs. 250 per share going-in price instead of Rs. 275, our return becomes 14.6% going in (1.96x capital). See the table below for numbers at Rs. 250:

At 300 we make only a 10.4% IRR and 1.64x multiple. So obviously, the margin of safety reduces the higher up we go and increases lower the price. This leads me to conclude that between Rs 250 and 275, there is a chance of making significant returns with a high margin of safety.

The upside potential is high if non-voice revenues kick-in post 3G auctions, if Zain does reasonably well, if Mr Mittal is able to create greater value for every rupee reinvested in the business and if the tower business is “IPOed” at a higher valuation. If any / all of these add a 10-15% growth to the PAT the stock could trade at around Rs. 800 in 5 years.

Concluding Note

The urban Indian today is hopelessly addicted to his cell phone and is not going to stop using it any time soon. In fact he will probably use it even more, not just to talk but also to check mail, news, sports updates, transfer money, browse photos, follow social networking sites etc. As an investor in Bharti that would make me feel really good. Every time you talk to someone using Airtel the cash registers are ringing and I am likely to get 20% of it. So in a way investing in Bharti is no different from investing in a cigarette manufacturing company – it has consumers who are hooked to the product and who shell out cash at fixed intervals so that they can keep talking on their mobile phones. Every time you see someone talking on the phone there is a 25% chance that he/she is using Airtel, making the investor the ultimate beneficiary.

What’s even better is that it’s the market leader in the industry, it is led by Sunil Mittal who has a history of creating shareholder value and creating and managing a highly low cost and profitable business model. It has a brand that is recognized throughout the country and used by 25% of all cellular subscribers. Finally, it’s a company with high entry barriers and significant competitive advantages.

At the prices I mentioned above, especially in the Rs. 250-275 range, I am willing to bet 50% of my hard earned net worth in this one company. Right now the price may be around Rs. 290 but as investors we need to be patient and wait for the right price to pay. I would advise you to wait for the stock to come in investment range and add slowly to your position, as I will be doing. I think it’s only appropriate to end with a quote by Ben Graham:

“It requires strength of character in order to think and to act in opposite fashion from the crowd; and also patience to wait for opportunities which may be spaced years apart.”