)
Consider these two facts. One, only 10 per cent of the nearly 900 million mobile customers use data services. Two, mobile penetration in urban India has already gone beyond 140 per cent. From these two statements, you don't have to be a rocket scientist to conclude that if you want incremental revenues, it has to come from data usage, not voice.
It would seem a natural progression from 2G to 3G and LTE, or Long-Term Evolution, the next standard of mobile communication, but telecommunication companies, or telcos, are facing a serious problem - they do not have enough spectrum. And without spectrum how can you offer live streaming video without buffering or movie downloads in seconds? If global averages are anything to go by, Indian telcos are starved of spectrum - on average, each telco owns 18 MHz, which is less than half the global average of 50 MHz.
Operators like Vodafone, Bharti and Idea Cellular bought 3G spectrum in 2010 in the 2100 MHz band at a stiff price in the hope of bringing about the data revolution. Their failure, perhaps, can be gauged from the fact that a market leader like Bharti has only 5 per cent of its 200 million customers using 3G. And because they are not getting commensurate revenues, companies are wary of investing in more towers with 3G equipment.
The problem is 3G operators have only 5 MHz of spectrum in the 2100 MHz band. And that is not enough for good quality 3G services. The global norm for a decent service is double that amount. Some companies have bought spectrum in the 2300 MHz band hoping that the larger 20 MHz chunk they get in that band will give them an edge. But Reliance Jio and Bharti which have stakes in this band have realised that 2300 MHz has serious limitations - the band has problems reaching indoor users, leading to inconsistency in the quality of service. This hitch can be partly resolved but that would require the telcos to erect double the number of towers and to install indoor antennas in homes. Also devices compatible for this band are expensive (at over Rs 20,000 apiece) and their mass adoption would not be easy.
Many companies have hedged their bets by buying spectrum in the 1800 MHz band, traditionally used for voice. In the auctions in February this year, Reliance Jio and Airtel bought large chunks of spectrum in this band in key circles. Of course, some of this will be used to bolster voice capacity but a part of it can be used for offering LTE services. Globally, the 1800 MHz band is one of the most used bands for LTE, also called 4G, with devices available for as low as $50 (Rs 3,000). According to analysts, this is a prudent step that will allow telcos like Bharti and Reliance Jio to integrate the band with other LTE bands (900 MHz, 2300 MHz ) in order to provide data services at higher speed to customers.
The effort to boost their potentially remunerative high-speed data services will prod the telcos to bid big in the next spectrum auction slated in 2015. As much as 156 MHz is being offered in the 900 MHz band across 18 circles. Two incumbent operators in each circle will be surrendering their spectrum on the expiry of the 20-year licence period, but the companies returning the spectrum, including Idea, Bharti, Airtel and RCom - will obviously fight aggressively to retain as much as they can of this valuable spectrum which supports over 250 million subscribers. Failure will mean a painful process of shifting customers to other bands. This will not only impact services, but force them to make additional investments too.
The 900 MHz band has the capacity to cover a larger geographical spread, thus reducing the need to put up too many towers. It also provides good indoor penetration capacity in comparison to the 1800 MHz band and is, therefore, far more attractive to the telcos.
The 900 MHz band is also tantalising in another aspect. While the band has been used only for voice in India, it is used globally to launch LTE services. With over 246 devices priced less than $100 out there to cash in on the band's LTE capacity, spectrum in this band spells revenues for telcos. That is why experts feel that there will be four to five players (apart from the two fighting to retain their spectrum) who will contend for the two slots. In other words, there is likely to be a repeat of the 3G spectrum auction that saw prices hitting the roof.
The other resource that the government is planning to auction is 800 MHz. CDMA operators have been clamouring for more spectrum in this band to enable them to offer LTE services. Players like Sistema have already announced their intention to offer 4G services. With just two serious CDMA operators lining up, the two to four slots likely to be on offer should prove enough. Like the 900 MHz band, 800 MHz is also a band that has caught the global fancy for use as LTE carrier and there are over 40 networks across the world. The price of devices in this band too are falling from their current perch at around $250.
3G over 4G
Despite 4G's temptations, everyone agrees that the mass adoption of data in India will come through 3G services. A senior executive of Vodafone India points out that LTE, for the time being, will constitute a niche market propelled by high data users mostly in large cities.
But the problem here is that the government has still not put out a road map on providing telcos the required additional spectrum without which their future could be seriously compromised. The situation would become dire given that the companies have already invested over Rs 50,000 crore to purchase expensive airwaves.
As analysts point out, the additional spectrum will probably have to come from the defence services. The Cellular Operators Association of India (COAI) and some technology providers have offered a solution: the government could swap the 1900 MHz spectrum - which no company requires - with 15 MHz of defence spectrum in the 2100 MHz band. Another option is to carve out the resource lying unutilised with the defence services and put it on auction.
If the defence services are amenable, these steps could be probable solutions to the scarcity of commercial spectrum. But the bigger problem remains. Rajan Mathew, secretary general of COAI, which represents the GSM section of the business, says that the auction of just two 900 MHz slots will lead to cut-throat competition. "This will force the prices upwards as happened in the 900 MHz auction in Delhi and Mumbai," he says. "That is why we are pushing the government to sell more slots."
COAI has two demands. One is to give the GSM operators access to the slots (of 5 MHz) earmarked for expansion of CDMA operators in the 800 MHz band, also called Extended GSM. Mathew says that because most operators are deserting CDMA, it does not make sense to create a growth path for that technology. It also hopes that spectrum lying, mostly unutilised, with state telco BSNL will be put up for auction.
It is unlikely that BSNL will succumb to pressure by surrendering its valuable spectrum. Similarly, CDMA operators are opposed to the sale of slots reserved for them to GSM players. "Their attitude is to kill us and keep everything for themselves," says a top executive of a CDMA operator about the GSM companies. "This spectrum is key to our future growth. It is part of our contract with the government." The government appears to be veering to their point of view and has rejected the recommendation of the Telecom Regulatory Authority of India to give the 800 MHz band to GSM players. Intent on selling the band to CMDA players, it has now asked TRAI to fix a base price for auctioning the 800 Mhz spectrum.
Many CDMA operators are pushing for the quick opening up of new bands - 700 MHz, for instance, for which there is over 90 Mhz of spectrum available for auction. They argue that there is already an eco-system for LTE in this band and point to the over 275 devices in the market and the over 50 networks using 700 MHz. But technology providers say that it is not a matured band and will take three to four years to reach stability when compatible devices will fall from their current steep prices of $400-500.
The government and the telcos know that the road ahead is not smooth. There is the need to take mobile broadband to the mass market, but it seems unlikely that the companies will get adequate spectrum that meets global averages. Will the data revolution then remain just a pipe dream?
