Like any business, we face the dilemma of outsourcing even in Telecom Industry. There are various views on outsourcing and the arguments of the proponents and opponents are equally convincing. Indian companies like Bharti Airtel, Idea, Vodafone Essar have outsourced some of the functions like network and IT which is still considered as core by many telecom operators in the western world.

Outsourcing is an important decision and it cannot be decided without giving proper thought to it. A framework is needed to help the operators decide on the contentious issue of outsourcing. If a function is core to the success of a company, then it cannot be outsourced but the problem is how do we identify the core. Geoffrey Moore explains the concept of core vs. context in his book “Living on the Fault Line” -

Everything begins with strategy. Strategy determines what is core to a company’s competitive advantage. Sustainable differentiation is the basis of economic success. Differentiation creates the basis for customer preference and gives a company pricing power. Sustainability is based on a barrier to competition and increases the returns on investment. Core is defined as any process that contributes directly to sustainable differentiation. Context is all other processes required to fulfill a company’s commitments to one or more of their stakeholders. Everything that gets done in a company is either core or context.

In optimizing resources, the goal in core is to create competitive advantage. Differentiation is critical here. This is the place to invest human and financial capital. On the other hand, with context activities, the goal is to meet market standards. Differentiating on context is a mistake and one that is costly. The key is to extract human and financial capital from context wherever possible and repurpose for core.

Moore uses the 2×2 matrix to determine the functions that can be outsourced but also cautions on the way it should be outsourced. I have taken the same matrix and applied on the Telecom Industry to arrive at the outsourcing framework for mobile operators (refer figure below).

The above becomes a very good framework to decide what needs to be outsourced and how. The companies cannot outsource the core that is mission critical. I would say that no operator can outsource its customers to any other company. Similarly, the development of employees cannot outsourced as it is the employees that define culture of the company and they bring in the customers. Strategy and Finance functions should always be kept close to the heart as they certainly help a company differentiate itself from others.

Anything in the top right quadrant (context and mission critical) is difficult to outsource but can be outsourced as they do not differentiate the business from competitors. Network operations were probably core when the mobile services were being rolled out but now it is no longer a differentiator (it is a hygiene) and hence can be outsourced. If we continue to insource network operations, we are likely to waste scarce resources and attention in an activity that does not give us any advantage. This may result in missing the opportunities due to lack of focus. The outsourcing of non-core activities also help us unlock the company’s wealth that can lead to higher stock prices.

IT services should also be outsourced as the companies do not have enough resources and cannot focus on development of cutting edge IT systems. Similarly, there could be other companies that can manage the VAS and customer support services better. One company’s context can be core of some other company. It is always better to outsource the context functions to a company that treat them as core. This would ensure that the value keeps generating within the system.

Functions that are support or non-mission critical and context can easily be outsourced. The challenge here is to ensure day-to-day quality and period quality checks should give the desired results.

For the context and yet mission critical quadrant, we need to have a very robust outsourcing strategy as the key challenge is keeping the control. The partner selection needs to be very carefully thought decision and the relationship should be governed by strong service level agreements (SLAs). In the next few articles, I would focus on how to outsource and what should be strategy of the operators while outsourcing.

References:

1. Geoffery Moore’s presentation on book ‘Living on the Fault Line’ - Geoffrey Moore – Core vs Context

2. Interview of Don Price, CTO-Bharti Airtel

If you liked this article, you may consider subscribing to Telecom Circle to get all the articles in your mail box

Recently, O2 announced a revised set of data tariffs for new and upgrading mobile customers in the UK. Customers will have a choice of “smartphone tariff plans” with a bundled data allocation of 500MB, 750MB or 1GB, depending on the total monthly fee, which ranges from £25 to £60. Earlier this month, AT&T had withdrawn unlimited data plans from its network. If the AT&T and O2 are any indicator of the future, the unlimited data plans could soon be a thing of the past.

In April 2010, Opera Mini users generated over 398 million MB of data for operators worldwide. Opera compresses the data by up to 90% and despite that this huge data was consumed by its users. Opera has close to 26.23% market share as per a report from Statcounter as of June, 2010. This means that the total data consumption across the world is much larger than 1150 million MB (Cisco estimates it to be closer to 2000 million MB) and this has been growing at over 100% annually. Cisco has predicted that the mobile data usage would continue to grow at over 100% CAGR until 2014.

If the data consumption continues to grow at the rate forecasted, the operators have a huge problem at hand. The increasing market share of iPhone and Android (Android has reached 10% share of smartphones in just 6 quarters) is likely to make the situation worse than Cisco’s forecast. In the figure below, it is clear that though Apple and Android have just 25% share in smartphone sales, they consume almost 67% of the total data traffic. This means that the average data usage on iPhone and Android based phones is a little under 3 times than that on any other phone.

The data networks were hugely under-utilized before the launch of iPhone. Operators had invested huge money in 3G networks and the due to low usage, they were finding it difficult to recover the investments. In order to increase usage and adoption, the carriers started to offer flat data plans and some of them went a step further by giving unlimited data plans. This was a great strategy on part of carriers and the analysts lauded it profusely and even lectured the carriers that were not offering unlimited data plans on its value proposition. With the advent of iPhone, the mobile Internet usage shot through the roof as the applications became more data intensive and the users started to download full track music, use peer-to-peer (P2P) file transfer and streaming services.

In US and some European countries, iPhone has really high market share putting extra strain on the carriers in those countries. O2 revealed that less than 0.1% of its subscribers account for a third of all network data traffic. Just 3 percent of users on smartphone tariffs account for 36 percent of its smartphone data traffic. The disproportionate data network usage by smartphone users (especially iPhone) meant that the other users were subsidizing the data usage of smartphone users. This is not only unfair but also unsustainable. O2 has been spending around £1m a day to upgrade its network to cope with the “exponential demand” for data on smartphones. AT&T had also claimed similar numbers on data usage.

It is getting increasingly clear that the operators in most of the countries would be left with two choices: either increase the data capacity by investing in the newer technologies like LTE or stop offering unlimited data plans. Given the financial health of the carriers and the maturity of LTE, it is likely that the carriers would adopt the later approach, i.e. stop offering unlimited data plans to consumers. Alternatively, the carriers can adopt the approach of promoting and offering incentives on handsets like Nokia and RIM that are either more data efficient or the data usage is lower by consumers on these phones.

Note: There is a poll embedded within this post, please visit the site to participate in this post's poll.
If you liked this article, you may consider subscribing to Telecom Circle to get all the articles in your mail box

Apple WWDC 2010 – Steve Jobs Keynote Quicknotes (Contributed by Navdeep Manaktala)

iPhone 4

• US Smartphone Market Share – RIM: 35%,iPhone: 28%, Windows, 19%, Android 9%, other 9%
• US Mobile Browser Usage – iPhone : 57%, Android : 23%, RIM : 13%, Others : 7%
• Hardware
  • New design (leaked iPhone)
  • Thinnest smartphone ever (claim) – 9.3mm thick; 24mm thinner than the iPhone 3GS
  • Stainless steel body for strength
  • Glass display for optical quality and scratch resistance
  • Integrated antenna – The outter rim of the case is the phone’s antenna
  • Display :
    • 3.5 in display
    • 960×460 pixels (4x more than current iPhone)
    • Retina display – 326 pixels per inch (300 is supposedly the highest for the human retina). Highest ever resolution on a phone
    • 800:1 contrast ratio (4x better than current iPhone)
    • IPS Technology for superb color and wide viewing angle
    • Going to set the standard for displays over the next few years (claim)
• A4 processor (same as on iPad. Designed by Apple)
• 40% improvement in battery life (larger battery + improved processor) – 3G Talk (7 hrs), 3G Browsing (6 hrs), WiFi Browsing (10 hrs), Video (10 hrs), Music (40 hrs), Standby (300 hrs)
• 32 GB of storage
• Quad-band HSDPA/HSUPA
• 3-axis gyroscope, adding pitch/roll/yaw, rotation around gravityvide 6 axis motion. Gyro + accelerometer pro – Perfect for gaming
• Camera : 5 Mpx, front & rear, backside illuminated sensor, allowing more light to get to the sensor, 5x digital zoom, LED flash, tap to focus
• Video : HD (720p @30 fps) video, tap to focus, built-in video editing, one click sharing, flash support for videos also
• iMovie App : Create video clips with ability to add pictures, backgrounds, location, music, etc.
• Software (iOS4)
  • Folders, task switcher, unified mailbox, enterprise features (as outlined earlier this month)
  • Choice of Bing for search
  • Will ship 100 millionth iOS device (iPhone, iPod, iPad) this month
  • iBooks now also comes to the iPhone (including download same book for all your apple devices at no extra charge). iBooks will also automatically sync your notes, books, and place wirelessly across all of your deivces
  • PDF support
  • Support for video calling (being called Facetime), front and rear camera support, only via Wifi in 2010, support for landscape and portrait mode
  • To be also made available for existing devices
  • Colours : Black, White
  • Price : $199 for 16GB, $299 for 32GB ; iPhone 3GS (8 GB) is going to be $99
  • Availability : Jun 24^th

iAds

• Objective – To help our developers earn money so they can continue building apps
• Selling iAds for 8 weeks now
• Adverisers so far – Nissan, Unilever, AT&T, Citi, Chanel, GE, Liberty Mutual, State Farm, Geico, Campbell, Sears, JC Penny, Target, Best Buy, Direct TV, TBS, and Disney. Have committed $60 million in iAd advertising
• Support for full screen and full interactivity

Platforms

• Apple supports two platforms – HTML5 (fully open, uncontrolled platform) and the AppStore (curated platform)

iPad

• 8500 native apps
• 35 million app downloads over last 65 days = 17 apps per iPad
• 5 million book downloads from iBooks over last 65 days = 2.5 books per iPad
• New feature on iBooks – Notes (post mark shows alongside wherever you noted)
• PDF support

AppStore

• 225k apps
• 15k apps submitted each week in 30 languages
• 95% of apps approved within a week
• Top reasons for rejection of submitted apps – App doesn’t function as advertised, Use of Private APIs & App crashes
• eBay app – 10 mn downloads of app since its launch last year. 600 mn transcations via the same in the first year and expected to do 1.5-2 bn transactions this year
• NetFlix coming to the iPhone (14 mn subscribers). Like on the iPad, you can resume movies from where you left off on other platforms. Full catalog and Instant Queue access
• Farmville by Zynga, one of the most popular online games of all time with 80 mn players currently, comes to the iPhone with pinch/zoom, notifications, etc.
• Guitar Hero with strumming (certain notes are swiped, rather than tapped) comes to the iPhone
• 5 billion downloads on iPhone. Across the 3 stores, there have been over 16 billion downloads
• Across all three stores (App stores, iTunes, iBooks), Apple has over 150 million credit cards users
• US$ 1 billion payouts to developers to date (70% of end user revenue)

Watch the video of the address on the Apple site.

There is a strong rumor of Qualcomm partnering with Airtel to roll-out TD-LTE network in India. Qualcomm has bid for the ongoing BWA spectrum and would need to find an Indian partner if it were to win the bid as Indian laws prohibit any foreign company to hold more than 74% share in any telecom company.

Qualcomm has a good chance of winning the bid as it would do anything to get the spectrum and showcase the TD-LTE technology. Airtel has not been able to get 3G spectrum in almost half the circles and would be keen to leapfrog technology by launching 4G across the country. 3G spectrum can then be used for freeing up the 2G network. Qualcomm on the other hand has no interest in launching the network by itself and is only interested in blocking WiMax from getting further foothold in India. Thus there is a perfect match between Airtel and Qualcomm.

I had mentioned in my earlier post that in a statement, Qualcomm made it clear that it intended to act only as a facilitator: If it wins the spectrum auction, Qualcomm plans to partner with an India-based operator to build a TD-LTE network and then exit the business. Surely, Qualcomm would like to partner with a pan India player and with significant subscriber base. Qualcomm has good relationships with Tata and Reliance but does not view them as potential partners which leaves is it with Airtel as the only choice. Vodafone is a global company and would not like to have a tie-up with Qualcomm.

Please note that this rumor has not been confirmed by either Airtel or Qualcomm.

Lately, a new dimension has been added to the fight between WiMax and LTE, which is TD-LTE. Developed largely in China, TD-LTE has started to make inroads into other markets as well. Huawei recently launched TD-LTE network for China Mobilewhile Qualcomm has put in a bid for BWA spectrum in 2300 MHz in India and plans to roll out TD-LTE network. It is reported that TD-LTE networks may also come up in Indonesia. In the US there have been statements by Clearwire that they want to migrate from their current Wimax to TD-LTE deployment. Reports state Russian operator Svyazinvest has also picked TD-LTE for mobile broadband deployment. GSM association has also thrown its weight behind TD-LTE. All in all, TD-LTE has crept out of its Chinese comfort zone and been unexpectedly linked to some of the world’s most influential mobile markets. These developments are a major blow to the WiMAX camp, the technology normally associated with TD spectrum bands.

What is TD-LTE?

There are two versions of LTE. FDD-LTE uses the FDD (Frequency Division Duplex) paired spectrum with two separated channels, one for the uplink and one for the downlink, which is the type of spectrum most mobile operators have. TD-LTE uses TDD (Time Division Duplex) unpaired spectrum channels that combine uplink and downlink, and split resources on the basis of real-time demand. Voice is inherently symmetric in the uplink and downlink so it is well suited for FDD spectrum allocations. Data traffic benefits from TDD spectrum, as it is typically asymmetric but the degree of uplink/downlink asymmetry is not fixed. The development of TD-LTE was initially pushed by China Mobile and regarded as a mainly Chinese standard, similarly to TD-SCDMA.

TD-LTE provides backward compatibility with 2G and 3G networks and hence would work even outside the LTE coverage area.

Why sudden interest in TD-LTE?

All vendors are trying to increase the installations of the network in which they hold the maximum number of patents and not necessarily moving towards development of a standard platform with a clear migration plan. In a statement, Qualcomm made it clear that it intended to act only as a facilitator: If it wins the spectrum auction, Qualcomm plans to partner with an India-based operator to build a TD-LTE network and then exit the business. The bulk of Qualcomm’s sales come from technology licensing agreements. Qualcomm has a veritable monopoly on CDMA patents and is seeking to expand the market for its other intellectual property holdings as the global wireless market moves toward 4G technology. The CDMA evolutionary path is coming to end after EVDO and hence Qualcomm would not like to risk the revenues by allowing the operators to move to a technology (read WiMax) where it has fewer patents. NSN, Motorola, Sequans and other companies are also reported to be active in the TD-LTE space so that they do not miss the bus in China.

In many countries, there are upcoming spectrum auctions for 2.3 GHz and 2.5 GHz spectrum bands that is better suited for TDD. Operators so far found little interest in the TDD band and WiMax had positioned itself very well for the TDD band. Operators are in general opposed to the WiMax roll out as it encourages open ecosystem and moreover was an entirely different technology with no evolutionary path from either GSM or CDMA. WiMAX  is controlled by IEEE, the consumer electronics industry, which is far more open than telecom/3GGP. However, now operators see a clear synergy in rolling out TD-LTE networks if the TDD spectrum comes up for auction. In India, the operators may actually be bidding for BWA to roll-out TD-LTE rather than WiMax.

There is another belief that is gaining traction is the fact that TD-LTE provides a migration path for WiMax players. Bruce Brda, Motorola’s senior vice president and general manager of home and networks mobility, said in an interview with a leading technology magazine that

Carriers can choose to deploy LTE networks either in FDD (frequency division duplex) or TDD (time division duplex) versions. Since WiMax is a TD technology and shares more assets with the latter, TD-LTE presents a more efficient migration option for WiMax operators. Those with broad spectrum rights such as Clearwire in the US have the option of dividing up that spectrum between WiMax and TD-LTE so that they can enter the LTE market without cutting off their existing subscriber base. But a small carrier with limited spectrum will not be able to maintain both business lines, and will have to decide between the two platform.

On the handset side, there is a flurry of activity with HTC announcing TD-LTE handset pilot and RIM’s announcement on TD-LTE Blackberry for China Mobile. Efforts are on to produce a converged LTE device, capable of supporting both TDD and FDD versions of the next-generation technology.

TD-LTE seems to be a technology which is seeing the ecosystem building around it and has a potential to co-exist with FDD LTE. It would be interesting to see if it indeed leaves behind WiMax in terms of roll-outs.

Recently read about the dynamic pricing plan from Uninor offering up to 60% discounts to its customers on the call costs based on the location and the time when the call is made and and the discount increases in areas with lower network traffic and during off-peak hours . Visit the link to get more info on the plan from Uninor.

Obviously Uninor is not the first company in the world which offered this kind of plan. Many emerging market operators have introduced this plan in various countries and many more are considering to do so. It is believed that this will result in the increase in the MOU and also bring down the burden on the networks where they are really clogged. MTN Group was the first operator to introduce dynamic pricing with its MTN Zone tariff in Swaziland and South Africa in February 2008, and has since rolled it out to 11 of its 21 operations. Vodacom South Africa quickly followed suit and introduced its own dynamic tariff, Yeboforless, in May 2008. More recently, Orange Botswana launched its new prepaid dynamic tariff (Sesolo Zone) in July 2009 and Safaricom Kenya announced the launch of its new prepaid dynamic tariff (Supa Onega) in September 2009. Orange Group has indicated that it is looking to trial a dynamic tariff offering in most of its African subsidiaries.

There are lots of analyst views (including Mohit’s) which say that it has the potential to increase ARPUs, MOU and hence the profitability for operators. They have always taken these plans as positive for the company. There are real examples which show that for a brief period of time, these plans do increase ARPU. One of them is Vodacom – ARPUs increased two consecutive quarters after the launch of dynamic pricing plans. I personally don’t believe that ARPUs increased because of this plan and there was something else which led to increase in ARPUs. The reason is ARPUs declined for the next 2 quarters after the first 2 quarters of increase and then increased once again after the decline. So ARPUs dont really follow a pattern apart from suggesting that ARPUs in Quarter 3 and Quarter 4 are normally higher than other quarters. Some of the other reasons why I believe this kind of plan is rather negative for the company are stated below. I know I am going against the collective intelligence of various operators but I have my reasons for the same.

• These discounts are based on the premise that these discounts would lead to uniform level of usage through out the day well distributed between high and low loads. However, in emerging economies where price elasticity seemingly is high, the introduction of dynamic pricing will lead to considerable change in the user behaviour leading to non uniform usage behaviour through out the day eroding price per minute. This risk could be minimized by having a real variable pricing plan where there are no preconfigured prices for any time and location. In fact pricing would really change based on the network load and not on time and location (I know network load is directly proportional to time and location of the user but still it matters when the prices are configured in the system)

• These discounts are aimed at cannibalizing operators own existing customer base by offering a blanket plan to the whole customer base. I think if operators need to offer this plan, then this should be offered to properly segmented customers based on their usage behaviour. Blanket plan would lead to lesser revenues from even that customer segment which is willing to pay full rates for the call during that time at that location. But by offering this plan to everybody operators can loose out on massive revenues. So plans should be launched after carefully studying the customer usage behaviour

• If operators want to uniformly distribute the network through out the day, then instead of cannibalizing their existing revenues, they should focus on other mechanisms/business models by which they can sweat their infrastructure by other means instead of loosing out on revenue from their customers who would change their behaviour. One of the most common businesses which comes into mind is M2M where ARPUs are really low and operators can charge least possible rates for these transactions promoting M2M transactions and usage. M2M can be automated in such a way that they take place only those times during the day when network usage is least. This would help operators not only capture very low ARPU business but also refrain from making less from those call where they could have made more. Dynamic pricing can seriously cannibalize operators own market.

• Could this lead to bad customer experience – With ever changing prices, customers could get really get confused about the cost of the making a call. If they made the call in one area and then moved to another, what price would they be charged. How many times in a day this pricing is changed. What are the T&Cs in these. These are the valid questions which operators will have to address otherwise customers could have a real bad experience dealing with this complex plan.

Finally, not to say that these plans invariably loose money but they have the potential to backfire and can lead to even lower ARPUs for the operators. What operators need is development of new business models to make effective use of their networks rather than dynamic pricing to cannibalize their own market

If you liked this article, you may consider subscribing to Telecom Circle to get all the articles in your mail box

In one of my previous post on 3G license fair price, I had written that I expect the Pan India 3G license to cost around $2 billion. However, now that 3G auctions are over the final price of a Pan India license is $3.6 billion ($1 = Rs 47), which is much higher than the my expectations and the expectations of all the analysts. The Government managed to garner $14.4 billion which is expected to bring the budget fiscal deficit down to below 5%. With the final price of $3.6 billion for a pan India license, it is now time to look at the options in front of the winners.

The results and winners of 3G auction can be downloaded here (3G_Auction_Final_Results).

The interesting thing about the auctions is that no single operator has bid for the Pan India license. Due to the high auction price, they have bid for the circles where they have strong 2G positions. Below is the interesting 3G network map of winners.

From the 3G winner map, it is clear that most of the operators have tried to bid for their stronghold circles except for Reliance and Tata. Airtel’s largest circles are Delhi, Karnataka, Andhra Pradesh, Bihar and Rajasthan where it has won the 3G licenses as well. Similarly, Idea has got 3G licenses for its strong circles like Maharashtra, Kerala, Haryana and Andhra Pradesh. Airtel and Vodafone have tried to ensure that their bid amounts for their key circles remain below the annual revenues, e.g. in Karnataka, Airtel has to pay $351 million which is just 49 per cent of the income generated there. On the other hand, Bharti dropped out of the race in Maharashtra and Gujarat where the bid amount had crossed the company’s annual revenue numbers. Clearly, the aim of the operators was to protect their current strongholds and current revenues. The figure below clearly shows the operator strategy was to defend their 2G revenues even if they fail to cover the large parts of the market.

The article from Economic Times on the relative competitive scenario post 3G auctions can be downloaded here (3G Economic Times)

What is the break-even user base for operators?

I have done a detailed analysis on the number of 3G subscribers required by operators in next 5 years to break-even in 5 years. The result of the analysis shows that the break-even subscriber base is highly sensitive to the ARPU that the operators can get from 3G subscribers.

The key assumptions in the above analysis is that the number of cell sites required across India would be around 75,000 resulting in a network CAPEX of $1,700 million and additional OPEX of $ 112 million per year. With the current ARPU levels of around Rs 150 ($3.2), I would not expect the 3G ARPU levels to be above Rs 350 ($7.5) which means that the industry would need to add over 52 million 3G subscribers in the next 5 years. If the 3G ARPU is lower than Rs 350, then the required user base for break-even goes up significantly. Given the current 2G congestion for large incumbents , the operators would endeavor to shift as many users to 3G as possible. Operators cannot price the 3G services high otherwise not many users would shift to 3G leaving the 2G networks no less congested from the levels that they are today.

What can be operators strategy to break-even on 3G?

Operators need to be very careful in the selection of their strategy options as the stakes are very high now. They need to work with the other ecosystem players to ensure that the total cost of ownership is as low as possible and that the relevant content is available to attract users to 3G. Here are some of the actions that can be taken by Indian operators to break-even faster than what most of the analysts think:

1. Replicate Minute Factory Model: The Indian operators have been innovative in bringing the costs down in 2G by changing the measurement metrics from ARPU to margins per minute. They have considered their business as a “Minute Factory” where minutes are sold at a certain price and there is a cost to the minutes. As long as the the realized rate per minute is higher than cost per minute by 30-35%, they are okay. Their entire effort has been to bring down the cost of minutes and have looked at network outsourcing, lower tariffs among host of other things. One of my previous posts on the case study on Indian operators has looked into the various actions taken by operators to keep the tariffs low. Even in 3G, Indian operators would need to follow the same “Minute Factory Model” in their efforts to attract higher number of 3G users. However, this action is likely to result in lower ARPU which would in turn mean higher 3G users required to break-even.

2. Refarm 900 MHz Spectrum: Unsuccessful bidders in any circle have an option of refarming their 900 MHz spectrum for 3G usage. This is an attractive option as most of the unsuccessful bidders have excess 2G 900 MHz capacity in the circles where they could not win. Refarming is a highly cost effective way of launching 3G services.  Refarming was explained in one of the previous posts on Spectrum Refarming – Rollout 3G services on 2G spectrum

3. Indirect bundling of handsets/ Upgrade schemes with handset vendors: Handset bundling in India is not prevalant as the ARPU levels are low which means that the handsets costs cannot be recovered even in two years time. Also, the operators have focused on keeping the costs low and hence have not indulged in any kind of handset subsidy. However, this should not prevent them from looking at innovative ways of indirect handset bundling. They should be willing to offer network minutes for free in return for tie-ups with handsets companies. The operators should tie up with handset vendors to upgrade the handsets of its subscribers who are on the verge of replacing their handsets by proactively targeting subscribers with over 18 months old handsets. Studies have shown that after upgrading their handsets, the users tend to experiment more with mobile services resulting in higher ARPU.

Handset vendors should also work with the operators to keep the aspiration levels high as well as keep the 3G handset prices low. There is a need to develop handsets that are “Made for India” keeping in mind the operator objectives. If the operators want to focus on voice, then there is a need for a cheap handset which may not have a big screen size or multimedia capabilities. On the other hand, if the operators decide to focus on music and videos, then the handset capability needs to be changed accordingly. Similarly, if the operators refarm the 2G spectrum then the handset vendors should have the handsets ready to support it. MediaTek’s plan of bringing cheap 3G smartphones based on Android might just be the impetus that this market requires. However, the other handset players need to better the MediaTek offering as the MediaTek based handsets may not be good for the ecosystem players and consumers in the long term.

4. Focus on Non Voice Devices: Operators should aggressively focus on non voice devices like the data cards, net books and other devices needing connectivity. This would ensure higher revenues and faster break-even. In the coming years, the popularity of net-books, eBook readers and handheld tablets is bound to increase and hence the need for connectivity.

5. Ensure fair Revenue share: For the success of 3G, it is important to have the right content and applications for the users. The ecosystem would be more vibrant if all the players get a fair revenue share. Fair revenue share would ensure higher developer interest in developing new applications. Unlike the other markets, the revenue share in India is heavily skewed in favor of operators which needs to change for quality content to be develop and mobilized.

6. Aggressively focus on GPRS: Once the users begin to shift to 3G, the 2G network would get decongested and the operators would be able to offer GPRS/EDGE plans to their subscribers. Subscribers should first experience internet and then would demand better speeds. Hence, GPRS can be a good stepping stone to complete 3G transition. It is therefore important for the operators to continue to focus on increasing GPRS penetration.

7. Roll-out HSPA+: India is already one of the last countries to roll-out 3G and therefore, it would be good for the operators to provide superior consumer experience in terms of internet speeds by rolling out HSPA+ rather than just WCDMA.

I am very sure that the operators would be able to break-even on 3G networks within 5 years if they follow the same that they did for 2G. India has always been different from other parts of the world in telecom and there is no reason why it cannot give positive surprises even on 3G.

Please leave your comments on the other possible strategies that the operators could adopt for faster break-even in 3G
If you liked this article, you may consider subscribing to Telecom Circle to get all the articles in your mail box

Uninor, which is a joint venture between Norway’s Telenor and India’s Unitech, came out with an innovative tariff plan to attract subscribers in a highly competitive Indian market.  In this market, competition intensity has meant lower tariffs but Uninor has introduced the dynamic pricing where customers would be offered 5-60% discount based on traffic on the network. The traffic would be function of location and time at which they call. The discount would change on an hourly basis and the discount available at any moment will be visible at all times on the screens of handsets that display cell broadcast. When the call ends, a flash will appear on the phone screen indicating the actual discounted cost of the call. Discounts are calculated by a sophisticated ‘Discount Engine’ that relies on state-of-the-art IT systems to continuously monitor traffic at every tower in the network. This plan is called ’24X7 Badalta Plan’ which in English means ’24X7 Changing Plan’.

Many analysts have dubbed this plan as a revenue depleater. They suspect that the effective call rates would fall and this is another way of triggering tariff war. However, I differ in the assessment and feel that this plan has a potential to increase usage. Anybody who needs to call would call irrespective of the time and place but if the user sees 50% discount flashing on the screen, he would be tempted to call-out to his friends or family. I would not be surprised if this plan leads to overall increase in ARPU due to higher minutes of usage.

Uninor has already outsourced its network and in case of managed services, the payment is made on peak capacity. This means that the increased utilization in off peak hours is at no extra cost till the time it does not cross the peak capacity. This effectively means that any increase in usage would go straight to the bottom line and hence the EBIDTA of this plan would be very high. If Uninor customers do not make call in the peak hours and wait for the discounts then Uninor pay outs would come down on account of lower peak capacity in future. This way it is a win-win situation. Operators have tried to give incentives for usage in non peak hours like night usage at half price but the pricing has never been this dynamic.

Another benefit of this plan would be ability of the operator to offer differential pricing in rural and urban areas. The capacity utilization is low in rural areas. The dynamic pricing of Uninor would offer higher discount in rural areas and lower discount in urban areas. This would not only help fill up the rural network but also help Uninor differentiate itself from other operators in the lucrative rural market.

The location based pricing is a radically new concept and it would be wrong to dismiss it as yet another means of lowering tariff. Having acquired the ability of monitoring the network utilization by tower on a real time basis, Uninor would be in a position to collect consumer data on the price elasticity which would help it in designing better plans. Most of the operators struggle with questions around price elasticity when it comes to plans aimed at increasing usage. I for one would be interested in tracking the success of this plan.

3G auctions finally took-off on 9th April after many delays and postponements. Government expects $7-8 billion from the 3G and BWA auctions and if the last 15 days are any indications, then the collections from the auctions are going to exceed the expectations. As on 27th April, the pan India license if for a premium of 147% at $1.92 billion (base price was $782 million). The auctions in all the circles would remain open till there is excess demand in any of the circles to maximize Government’s revenues. The question to be asked is where would the carriers stop bidding and what is the fair value for a pan India license?

Background:

There are around 14 operators in each of the service area in the country. The high competitive intensity has led to a tariff war with the realized rate dropping below 1 cent/min making India the cheapest country in the world when it comes to mobile telephony. Out of 14 operators, almost 50% are new operators which are still in the process of launching 2G networks and have a combined market share of less than 10%. The presentation below provides a good overview of the Indian Telecom Market.

The Government of India started the process of auction of 3G and BWA airways on 9th April with an objective of garnering $7-8 billion and has set the base price of a pan India 3G license at $780 million. There are 4 slots available in all the circles except for Punjab, Bihar, Himachal, J&K and West Bengal where there are 5 slots available. One of the slots in each circle has already been awarded to the state owned company, BSNL (MTNL in Delhi and Mumbai) so effectively there are 3 slots for auction in most of the circles.

What are the analysts estimates?

Many of the top equity analysts and investment bankers have estimated that the operators are going to bid around $2 billion for a pan India license. The table alongside shows that analyst views on the 3G auction price for pan India license. The analysts have based their claims on high data revenues from 3G.  Motilal Oswal has assumed non voice revenues of 20% of total revenues by 2015 and 12% annual growth in wireless revenues for next five years. In the last few quarters, due to competitive tariffs, the operators have seen negative revenue growth and I do not see the wireless revenues to grow at 12% CAGR for five years. Similarly, BNP Paribas has assumed the data ARPU to be Rs 100 which is very high considering it is close to Rs 18 currently. Is there any killer 3G application that would increase the data ARPU? I do not think there is any application that cannot run on 2G networks. No doubt the experience would be much better but the country is yet to adopt the mobile internet. So expecting the data usage to shoot up significantly with 3G would not be right.

The latest 3G auction bid value can be accessed from the DOT Website.

Will the operators buy the analyst argument?

They have no choice but the get the spectrum. There are four serious bidders for the pan India license and there are only 3 slots available. The cost of not getting the 3G spectrum is high due to the following reason:

1. High congestion: In most of the metro and A circles, the networks of the leading operators are choked due to spectrum constraints. The operators are likely to use the 3G spectrum for voice thereby reducing congestion on 2G airways. The availability of higher amount of spectrum would also help the operators reduce their capital expenditure (CAPEX). In my opinion, the operators are not bidding to launch high speed networks but are bidding to get as much spectrum as possible.
2. Service Differentiation: Currently, there is very little differentiation amongst the operators and with the mobile number portability (MNP) round the corner, no operator can miss the opportunity of becoming a differentiated player using 3G.
3. Higher Data Usage: Operators are hoping that better experience on mobile internet is likely to drive the data revenues. With the 2G network becoming less congested, even the GPRS experience is likely to be much better. I believe that there is a potential to increase the data revenues to 15% of total revenues in the near term but not to the levels that some of the analysts are projecting.

What would be final auction price for pan India license? It is clear that no big incumbent operator can afford to be bow out of the 3G and they will try to get the spectrum at any cost till the time their balance sheets are able to support the license fee. The operators would need to fund the 3G license fee as well as fund the rollout of 3G networks. This would result in higher debt on their balance sheets leading to higher debt to equity ratio (leverage). This means that the ability to fund the 3G license and the networks would vary from operator to operator depending on the current levels of debt. As per the analysis of BNP Paribas, the debt to equity and debt to EBIDTA levels pre and post 3G license acquisition would be as follows (assuming $2 billion as the cost of 3G license):

Bharti has negligible debt but after the acquisition of Zain, its balance sheet would soon get $10.7 billion as debt but still it would be in a better situation than Reliance and Idea to be able to fund the $2 billion 3G license cost. In case of Reliance and Idea, the net debt to EBIDTA becomes abnormally high which makes me believe that at some stage they would walk out of the bidding for pan India license and would concentrate on getting the licenses in circles where they are strong. Vodafone and Tata Docomo would be able to fund their 3G aspirations due to availability of funds from Global organization for Vodafone and from Docomo for Tata. This means that the fight would be amongst Airtel, Vodafone and Tata to get the pan India license but the other players especially Reliance and Tata are going to drive the price in a few circles to an unsustainable levels. This means that the pan India license would finally cost around 2 billion to the operators.

Is $2 billion fair price for 3G license?

It is difficult to say if $ 2 billion is the fair price or is the 3G license overpriced in India. It would all depend on the data take up by consumers and the operators ability to price the service in a way that they are able to maximize the 3G subscribers on a base station. This means it would be in the interest of operators to keep the tariff low to gain consumer traction and limit the geographic expansion till there is a strong business case to do so. This would help bring down the CAPEX. The analysts have been talking to the operators on the possible license cost and have been spinning the business case around the $ 2 billion mark by playing with assumptions around data uptake and ARPU/wireless revenues. I am sure the Indian operator would not be looking at data as the only driver but would apply the “Minute Factory” model to 3G as well. This would ensure mass adoption of 3G in the country as it happened in case of 2G.

The smartphones has been the fastest growing category within the mobile phone space for the last couple of years. In 2009, a total of 172 million smartphones were sold which was a growth of 24% over 2008 (source: Gartner) whereas the mobile phone category was more or less flat in 2009 in the backdrop of severe recession. Analysts expect over 240 million Smartphones to be sold in 2010.

What is a Smartphone?

Unfortunately, there is no one universal definition of Smartphones. Some analysts have defined it as phones using open operating system which means that Symbian, Windows and Android based phones qualify as Smartphones but then what about Apple iPhone and RIM that run on proprietary operating system? Some define it as phone that gives PC like functionality and for some, it is the one with most advanced mobile phone features and hence the feature set of smartphones keep changing. A few industry analysts define it at mobile phones running on operating system with standardized interface and that provide easy access to developers for application development.

Whatever is the definition, it is commonly accepted that Smartphones have strong email clients, third party applications of some kind, QWERTY hardware or software keyboard support, high-speed internet, powerful calendar, contact and organizational features and support for powerful processors and touch screens. The operating systems that can support the above feature requirements are Symbian, Windows, Android, iPhone OS, RIM, Maemo (Meego going forward with Maemo’s merger with Moblin) and Palm OS.

Why does everybody want to win the Smartphone battle?

Smartphones are not even 10% of the total mobile phones but still all the handset players want to win this battle for the following reasons:

1. High ASP - The average selling price of a smart phone is almost 3 times that of a non-smartphone which means it has a higher impact on not only the value share but also on the profitability. Apple and RIM though have just 3% market share in the mobile devices but have over 50% share in industry profits which reflects the high profitability of the smartphones.
2. Thought leadership - Success in the smartphone business gives the device vendor the status of “Thought Leadership”. Any vendor having the “Though Leadership” benefits in the non-smartphone devices as well as the consumers want to buy the devices from the leading brands (in terms of “Thought Leadership”) even if they are not buying the devices that gave the vendor this status. This status is like the quality assurance certificate.
3. Future of mobile phones – The smartphone market is expected to expand significantly in future to almost 40% of the total device market by 2013 which means that no vendor can afford to ignore this market.
4. Operators prefer Smartphones – The increasing popularity of iPhones has led to increased data usage on the mobile. This has enabled the carriers to maintain their ARPUs despite fall in voice tariffs. The data usage on smartphones is almost 3 times higher than a normal phone. 55% of iPhone users use mobile social networking and 80% use it for surfing web daily. The high data usage has led to the situation of data scarcity from data abundance and carriers love this situation.

Who will win the battle?

Anybody who wins in the smartphone business has to win it on back of the operating system.

Symbian has been the market leader in the Smartphone segment of the market with 47% market share in 2009. However, Symbian is facing a stiff challenge from iPhone OS, Android and RIM OS and has lost market share in the recent few quarters. The figure alongside gives Smartphone market shares in 2009. Android though has only 3.9% market share in 2009 but is making the most noise in the smartphone segment.

Each operating system has its own positives and negatives and before we attempt to answer the question that who is going to win the battle, it is important to evaluate the pros and cons of top operating systems:

Andriod:

This operating system has been developed by Google with the aim of not only to get a foothold in the lucrative mobile industry but also to change the way the mobile owners consume data on the net.

Pros – Open, free and supporting many devices. Great experience of Google’s web properties on the mobile like messaging

Cons- Fear of excessive fragmentation due to its open architecture and is also more susceptible to security threats

iPhone OS:

This operating system has been developed by Apple and is a closed, proprietary operating system and is only for Apple devices, iPhone and iPod Touch. iPhones have been so popular that they are now being blamed for the increasing data congestion on the networks.

Pros - Single unified platform and its biggest asset is the application store

Cons- Proprietary and hence its dependence on one vendor is a risk. Application submission process is a cumbersome and Apple keeps a tight control on the 3rd party applications.

Blackberry OS:

Blackberry OS is owned by Research in Motion (RIM) and is a proprietary operating system.

Pros - Strong in enterprise mobility segment and has created a perception amongst IT managers of being a much secured OS. Strong messaging service

Cons- Proprietary and hence its dependence on one vendor is a risk. Aged operating system and requires the special BES server. Lack of focus on consumer segment in the past limits the lure to potential developers. This operating system also suffers from lack of optimization on touchscreen devices and a bad web browser.

Symbian:

Symbian is owned by Nokia  and used by many other vendors like Samsung, LG and Sony Ericsson. It is market leader but is seen as an archaic operating system. However, it is still holding out with promise to launch Symbian 3 and 4 by the end of the year.

Pros - Largest installed base and hence the economies of scale. Backing of Nokia, the market leader helps Symbian maintain its market leadership.

Cons- Aged OS and not really optimized for touchscreen devices. It is virtually absent in the North American markets and is facing huge competition from Android, iPhone OS and Blackberry OS. Dependence on Nokia is a risk as many of the other vendors are shifting to Android.

Windows Mobile:

This operating system has been developed by Microsoft and promises to mirror the PC experience on the smartphone. After continuous decline in market share in the past many quarters, Microsoft recently announced Windows Mobile Version 7 which has got rave reviews from analysts and handset vendors.

Pros - Backing of Microsoft which has virtual monopoly on personal computers. Microsoft’s ability to provide resources and its possible integration with its other hot properties like X-Box, Zune can ignite developer interest in this operating system.

Cons- Past failures to haunt version 7. Heavy operating system and hence requires higher hardware specifications. Last version had a bad user interface (UI) and web browser. I hope the new version has targeted the UI for improvement

There are many other operating systems like Maemo, LiMo, Samsung’s Bada that are trying to get a foothold in the lucrative smartphone business but only time will tell if they can manage to break into the top five operating systems.

I am not going to pass a verdict on which operating system will win but I will publish the results of the poll that I conducted on this site and no prizes for guessing the winner (refer to the image on the left). The winner was Android with over 40% of votes. In the backdrop of its popularity and the confidence that the ecosystem players are putting in Android, it is not at all surprising that it is going to be the fastest growing operating system for next five years. In the recently concluded Mobile World Congress, Android managed to get the highest number of handset announcements. Though even iPhone OS is a wonderful operating system but its proprietary nature would limit its expansion.

In your opinion, which operating system and vendor is likely to win in the smartphone space? Please do comment and cast your vote.

Note: There is a poll embedded within this post, please visit the site to participate in this post's poll.

Our attention focuses on recent and upcoming announcements that peg events for rapidly evolving next generation wireless competition.  From the view of the author, among the many events, the top two events that stand out most are:

• Sprint-Clearwire’s announcement of the HTC EVO Android 2.1, 4G WebPhone
• Qualcomm’s intent to bid on the India auction following April 9^th for BWA with TD-LTE

Sprint’s CEO, Dan Hesse, is expected to unveil the HTC Supersonic, officially announced as under the EVO 4G brand name, a leading-edge SmartPhone that will be the first market-leading phone to work on both 3G and Clearwire’s WiMAX network during his keynote address and in press briefings at CTIA.

The EVO 4G is a successor of the well reviewed HTC HD2 model. Specifications include:

• 4.3″ TFT, (~ 1/3 larger than the iPhone)
• 3G/4G WiFi hotspot extension capability
• 8 MP Primary Camera with Dual LED; Auto focus, Digital zoom
• 1.3 MP front-facing camera targeted at BB hogging social networking, web-casting of pictures and streaming video
• Kickstand for stationary use incl. HDTV viewing
• 3.5mm Head Jack
• Clearwire WiMAX, Sprint 3G data network coverage
• Stereo Bluetooth
• GPS, proximity and motion sensors
• Digital compass
• 1 GB ROM
• 512 RAM
• 1,500 mAh battery
• 1GHZ Snapdragon
• MicroUSB and MicroSD card clot
• Micro HDMI
• Qualcomm 1 GHz Snapdragon processor with EVDO
• Sequans WiMAX chip
• Social networking with most popular sites
• Android market access to over 30,000 applications

The EVO 4G represents that the WiMAX ecosystem and deployments have crossed a long anticipated threshold as a competitive full featured converged network.  Clearwire and Sprint will look for entry of this and other subscriber-growth driving Android and Microsoft Mobile 7 devices that will be introduced over the next year to ramp sales sharply.

While it can be argued that Clearwire’s primary strengths lie in wireless broadband, more as an adjunct to other networks rather than as a mobile hand-held device juggernaut, the onus on 4G providers will be to fulfill a pallet of market needs.  The device category that has become the most subscriber grabbing and, therefore, greatest threshold for capturing mind and market share is the leading edge of the SmartPhone and larger form-factor MID devices.  Sprint-Clearwire takes the BB strengths of the NG network to deliver more fully on the promise of 4G: full-tilt-boogey broadband anywhere.

Of necessity, Clearwire’s first priority has been to build out a substantial network footprint.  Regardless of the value proposition of the WiMAX/3G combination, leading edge devices require  large markets to sell into. As Clearwire has reached around 50 million POPs currently and expects to cover over 110 by early 2011, the draw for handset providers has crossed a threshold.

The remaining details of the EVO 4G  are almost unimportant: it follows previous HTC units which have improved over time to become challengers to the iPhone, Motorola DROID, which is similarly based on Android OS. Price and availability of the device and service plans are keys to success.  If married to similar Sprint’s aggressive always connected package strategy, the EVO 4G may carve out a revival for Sprint and a surge in subscriber growth for Clearwire.

No ‘Lines in the Sky’

On the other side of the world, Qualcomm entered the fray for the week of April 9th BWA auction in India. This throws a new wrinkle into the development of 4G and causes reflection on the ‘Clash of the Titans’ between mobile wireless and ITC worlds:

• One of WiMAX’ primary tenets has been that standards should develop based on the merits of underlying technology
• Furthermore, spectrum should be regulated and used as technology agnostic
• IEEE 802.16, 3GPP, and related standards groups have evolved efforts towards use of IP for multiple services from embedded machine to machine, MtM, to high-end mobile devices
• As Qualcomm lost their gambit to promote UMB as a worldwide standard for 4G, they became a strong advocate for LTE with their IPR licensing position ostensibly in long term jeopardy.

The Qualcomm bid could have a number of possible repercussions:

1. Delay of use of a portion of the spectrum.  However, the auctions are arranged so that no bidder can block development in a specific geography.
2. Prying away of some WiMAX deployments to LTE, particularly the TD-LTE version spearheaded by the Chinese.
3. An increased market potential for WiMAX/LTE multiple mode chipsets.

Each of these events portrays how much the industry has converged: WiMAX is crossing the threshold to become more mainstream, even while limited by spectrum access. LTE a close cousin in terms of technology is being positioned as both a BWA as well as next generation mobile network.

Sprint and Clearwire’s recent interviews further clarify that the company is, after all, a network operator rather than a religious technology advocate.   Sprint’s choice of the name “EVO 4G” is particularly well suited and telling: one flavor of technology or the other does not shackle The ‘nature’ of devices on IP networks.  In fact, the major reason for the disruptive shift to open applications development environments and open ‘all-IP’ WRAN, wireless radio access,  has been to allow greater degree of freedom from specific hardware network and device environments.  The introduction of the EVO 4G marks a capstone event for the coming together of broadband and wireless.

Often there is a debate on the next form factor (physical appearance) for mobile devices. There are fancy form factors like fold-able, morph, etc. that are often quoted but I believe that the next wave would be of an existing form factor which is QWERTY.

QWERTY phones have been in use for a long time but they have been widely marketed by device vendors as messaging phones. The trend of linking QWERTY with messaging was started by Blackberry with their ugly looking devices that had the perfect email solution. Later on any QWERTY device was labeled as business device. However, this myth is changing very fast now with the launch of cheap QWERTY devices particularly by Chinese vendors. Chinese vendors, known for innovation, launched QWERTY devices at around $50 price point which turned out to be the tipping point for such devices. These devices were lapped on by financially constraint individuals who were not much literate as well. They found that using a QWERTY phone made the storage of contact details in the phone book easy. They could write SMSs without any goof-ups like writing “Porn” for “POSM” (POSM-Point of Sale Material) or “Pink Panties” for “Pink Panther” while using the predictive text. It helps simplify the input mechanism which is a big barrier to cellphone adoption. The point is that it was wrong to club QWERTY devices with email and social networking. I am not saying that email is not easy with QWERTY but I am trying to emphasize that the market for QWERTY devices would have been much larger had the marketers realized the utility of QWERTY for the poor and illiterate people.

I would reckon that in the coming 3-4 years, we would see growth in shipments of QWERTY devices as high as 200% every year with most of the demand coming from emerging markets like China, India, Africa. The low cost vendors have been able to catch on with this trend pretty early. Will the established players see things from a different prism?

The launch of Apple iPhone three years back made the touch screen popular in the mobile phones. Long back, before Apple, Nokia had flirted with touch screens but did not get the same success as Apple. Some people attribute the success of iPhone, apart from itunes, to its capacitive touch screen and its ability to support multi-touch. Suddenly, all the phone manufacturers wanted to have touch screens but most of the vendors could only supply resistive touch screens which according to a few people did not offer the same user experience. So what is it about capacitive touch screen that makes it stand out? Before I share my views on this, I would first explain the two types of touch screens and the difference between the two:

Resistive Touch Screen

Resistive touch screens are composed of two flexible sheets coated with a resistive material and separated by an air gap or microdots. When pressure is applied to the screen, by a finger or other objects, the sheets are brought into electrical contact which causes a switch closure and a command is registered. The figure (source: Screen Tek) on the left shows how does a resistive screen works. Examples of resistive touch screen phones are Nokia 5800 Samsung Omnia, etc.

Advantages:

1. Cost effective and duarble
2. Can work in all climate conditions and in any environment like the restaurants, hospitals, etc.
3. Input can be made using a stylus or finger. However, stylus is better suited.
4. Less sensitive to scratches as compared to capacitive screens

Disadvantages:

1. Normally does not support multi-touch though Nokia has announced that on its handsets, multi-touch would be supported on both resistive and capacitive touch screens
2. Recalibration might be required over time as resistive touch screens are known to drift over time. However, this is a simple process and can be performed by users by a simple dot mapping.
3. While using fingers, slightly more pressure needs to be put on the screen
4. Provides only 75% optical transparency which means that its clarity is lower than capacitive touch screen

Capacitive Touch Screen

A capacitive touchscreen panel consists of an insulator such as glass, coated with a transparent conductor (mostly indium tin oxide). As the human body is also a conductor, touching the surface of the screen results in a distortion of the body’s electrostatic field, measurable as a change in capacitance. Unlike its resistive counterpart, capacitive touch screens must be touched by a human finger or some sort of conductive device being held by a human hand. The figure on the right from Electronic Design shows how does capacitive screens work. Examples of devices with capacitive touchscreens are the Apple iPhone and the T-Mobile G1. Advantages:

1. Higher clarity display (up to 90% optical transparency)
2. Supports multi-touch
3. High touch resolution
4. High sensitivity

Disadvantages:

1. Needs a human finger to register input. It is not possible to use this screen wearing gloves
2. Cannot be used in all weather scenarios

Which touch screen is better?

Due to the success of iPhone, there is a common perception that capacitive touch screens are better suited for cell phones than the resistive touch screens. However, I believe that it does not matter whether the mobile phone has a capacitive screen or resistive screen – what matters is the software behind it. The operating system defines the user experience and if the operating system is not optimized for touch screen, even the capacitive touch screen would not enhance the user experience.

Do you agree with me? I am eager to hear your views on the issue.

Spectrum is a scarce resource and hence the operators scramble to garner as much as they can. In the coming few years, unprecedented amount of spectrum will be freed up in the switchover from analogue to digital terrestrial TV which is better known as the Digital Dividend. The spectrum is getting freed as digital TV is much more spectrum efficient than analogue TV.

The Digital Dividend spectrum is located between 200 MHz and 1GHz. This spectrum band offers an excellent balance between transmission capacity and distance coverage. GSMA claims that if just 25%, or around 100MHz, of the spectrum currently used by analogue TV (470 – 862 MHz) was re-allocated to mobile communications, the mobile industry could dramatically speed up the rollout of broadband communications and increase coverage. However, Analsys Mason, in its report to European Commission identified 5 potential areas for usage of digital spectrum

1. Digital Terrestrial TV
2. Broadcast Mobile TV
3. Commercial Wireless Broadband services, both to fixed location and mobile devices
4. Wireless Broadband services for public protection and disaster relief (PPDR)
5. Services ancillary to broadcasting and programme making (SAB/SAP)

In my opinion, the most appropriate use of freed spectrum would be to take the mobile broadband to the masses especially in the rural areas and far flung areas. There is a digital divide amongst developed and developing countries and w;ithin each country between the urban and rural areas. The excellent propagation capacity at low frequencies means fewer base stations are required leading to a cost effective network roll-out with wide and in-building coverage. It is estimated that mobile broadband in 2100 MHz band is 3 times more expensive than that in 700 MHz band (Refer the Capex chart below from GSMA).

Benefits of Mobile Broadband

There are several benefits of mobile broadband:

• Economic Development – there are reports that suggest that with every 10% increase in household broadband penetration, the GDP growth can go up by up to  1.4% (source: McKinsey). Bringing broadband penetration levels in emerging markets to today’s Western European levels could potentially add USD 300-420 billion in GDP and generate 10-14 million jobs.  Mobile broadband would help increase the broadband penetration faster as the mobile penetration is much higher than PC penetration.
• Mobile Broadband can be used for tele-medicine, education and general information to farmers, fishermen, etc. which would lead to prosperity in rural areas and among lower income groups. This is a direct result of productivity increase due to broadband usage.
• Development of human capital as internet can bridge the knowledge gap
• Mobile broadband is cheaper than fixed broadband as the last mile connectivity does not require laying of copper wires for mobile broadband. This means that not only the cost is less but also the deployment is faster.

There are many impediments to allocation of spectrum to mobile operators. Most of the countries are yet to make up their mind on the spectrum allocation, also the timelines vary across countries for switch over from analogue TV to digital TV. In Oct’07, the World Radiocommunications Conference (WRC) discussed the issue of allocation of digital dividend spectrum and agreed to identify a chunk of UHF spectrum for mobile broadband services, and a chunk of spectrum in the higher frequency bands to create the capacity required for the next generation of advanced mobile services. The following UHF band identifications were made at WRC 07:

Despite the WRC recommendation, there are a few countries that are planning to adopt their own distinctive band plans for the UHF spectrum rather than coordinating their spectrum allocations with other countries. Given the multiplicity of spectrum bands across regions, there is a need for harmonization of spectrum bands to keep the cost low. The key reason for success of GSM has been the standardization of spectrum bands which kept the terminal costs down. However, in case of mobile broadband, there are multiple frequency bands and to enable roaming, the handset would need to work on all or most of the frequency bands. Frequency harmonization can drive down terminal costs by as much as 50% (Refer the GSMA white paper on ‘The advantages of common frequency bands for mobile handset production’ )

In my opinion, the regulators across the world should allocate the digital dividend spectrum to mobile broadband for bridging the digital divide without any further delay. There is a need to keep the service and terminals affordable that can only be achieved if consensus is attained on allocating the 650-862 MHz for LTE. Given the scarcity of spectrum, it is important for Governments to taken the decision of switchover from analogue to digital TV and the mobile broadband can be used as one of the channels for TV.

On 15th Feb, 2010 Bharti announced the acceptance of its bid for Zain’s Africa business by Zain’s board. The deal value is $ 10.7 billion which Bharti plans to fund by loans arranged by Standard Chartered Bank and a couple of other Banks. This is the third attempt by Bharti to enter Africa after two failed attempts with MTN of South Africa. The share price of Bharti took a serious beating after this announcement as most of the analyst felt that Bharti is overpaying for Zain’s assets by around $ 2 billion. My analysis shows that it is a good deal for Bharti and its stock price would rise in the long run. In fact, just after the announcement of the deal, I brought Bharti shares and my optimism is based on the following facts:

1. Low Financial Leverage

Bharti has a very low “Net Debt to Equity Ratio” of 0.05 at the end of Dec., 2009 which means that it is virtually a debt free company. It is good to have low debt but zero debt is not a desirable situation as debt can increase the shareholders’ return on their investment due to tax advantages associated with borrowing. The figure along side illustrates how financial leverage can help companies optimize their cost of capital. As it can be seen from the figure, the total cost of capital is high if there is no or low debt and there is a level of financial leverage where the total cost of capital is the lowest. Most of the companies try to maintain leverage at the level where the total cost of capital is lowest.

Bharti is a profitable company with over 40% EBIDTA margins which is higher than the cost of debt. This means that it is better for the company to pay interest than paying dividends to a large number of shareholders and hence it should either reduce the shareholding (through share buyback) or increase debt and deploy debt in a profitable way. Bharti has selected the second option and is taking debt to buy Zain that would return higher profits in the long term. It is like investing for the future. Even if Bharti were to pickup $ 7 billion for the Zain acquisition as debt, then also the leverage is unlikely to exceed 1 which would still be lower than many listed companies. To put the case in perspective, Verizon has a debt to equity ratio of 1:45 while the similar ratio for Sprint Nextel is at 1.19 and for Telefonica, it is as high as 2.78 (source: Forbes financial application).

2. Free Cash

Bharti is one of the few carriers across the world that has free cash flow and it does not make sense for the company to keep sitting on the pile of cash when it can deploy it in productive assets. The capex in the Indian operations have started to decline and hence the free cash flow is likely to increase even further in future. If Bharti decides to fund the Zain acquisition through debt, it would not find much problem in servicing this debt due to generation of free cash flow in the years to come.

3. Attractiveness of African Market

Africa is the next frontier as far as mobility is concerned. As markets saturate, the carriers start to look at opportunities outside of their domestic markets. Vodafone invested in India around three years back to increase the growth potential of its revenues. Now that the tariffs have declined significantly in India and that the penetration levels have crossed 45% in India, there is little opportunity left in the domestic market for Bharti. The penetration levels in Africa are around 33% and the ARPU levels are high varying from $8 – 12 (apart from Kenya and Ghana where it is closer to India ARPU levels of $4). Bharti can replicate its low cost model in the African market which would not only bring the cost down but would also result in significantly higher subscriber addition. The level of competition in Africa is not as intense as India as most of the countries have no more than 4-5 operators. The countries where Zain has operations in Africa have a population of close to 500 million which is an indicator of the opportunity that lies in Africa.

Summary

I believe that with this deal, Bharti would be able to increase the valuation of the company in the longer term. The Zain deal would turn out to be better than MTN as it would be an outright purchase rather than co-ownership as in case of MTN deal. Most of the joint ventures or mergers fail due to cultural issues and ego issues between partners and the ego issues are more likely in co-ownership model. Moreover, in case of Zain, the regulators are unlikely to hold the deal on account of nation pride as Zain does not belong to any of African country and hence there would not be any insistence on dual listing or any similar binding as it was in the case of MTN.