Accelerating 3G in India

Updated: May 2 2013, 09:15am hrs
Rekha Jain

If India can harmonise its 3G and 4G bands in keeping with what the rest of the world is doing,

the affordability aspect of mobile broadband services can be addressed

In India, like in many other developing countries, the ubiquity of mobile phones, sparse wire line infrastructure, high cost of PC and unreliable power supply make mobiles the dominant platform for internet and broadband access. Since mobiles and broadband are the key drivers of economic growth, the increasing gap between developing and developed countries should lead us, as Indians, to think about how to accelerate this deployment and adoption. Obviously, there are several steps that need to be taken.

In the 2G band, predominantly used for voice, services took off exponentially due to rapidly falling prices of 2G devices and services. This was because of competitive pressures and the large and increasing base of users, both nationally and internationally, that gave economies of scale. Nearly 75% or more of the mobile subscriber base used GSM services offered in the 900MHz or 1800MHz band, leading to harmonisation across countries. Since internationally there were two dominant bands, it was possible to have economies of scale, and a virtuous cycle of greater adoption and falling prices emerged.

If a similar harmonisation can happen in India with respect to international scenario in 3G and 4G bands, used for mobile broadband services, the affordability aspect can be addressed. For mobile broadband in India, two frequency bands have been so far allocated in the 2100MHz (3G) and 2300MHz (BWA or broadband wireless access). Of this, the 2100MHz band is internationally harmonised for 3G and increasingly 4G services. The other key bands where mobile broadband services are globally offered700MHz, 900MHz, 1800MHz, 2300MHz and 2500MHzhave limited deployments in a few countries. In India too there are regulatory bottlenecks in these bands. For example, the date of 700MHz for its refarming from broadcasting to mobile broadband is uncertain and the spectrum bands in the 900MHz and 1800MHz bands need to be refarmed for broadband services in the future. Therefore, 2100MHz band looks to be the most promising in the near term.

In India, a large part of the commercially-valuable spectrum has been with the ministry of defence (MoD). The Department of Telecommunications (DoT) has been working with MoD to offer alternatives to get such spectrum released. For example, a fibre optic backbone was to be provided by the Bharat Sanchar Nigam Limited (BSNL) in lieu of MoD vacating some parts of the 3G band. Delays in provision of the backbone led to a constrained availability in 3G bands as MoDs release of spectrum was also delayed. This resulted in a limited number of licences. Further, prior DoT policies of allocating spectrum based on ever-tightening norms of subscribers, and increasing uncertainty about the 3G auctions, created a situation of aggressive bidding and high auction prices. Consequently, no operator other than BSNL/MTNL got a pan-India licence. (While BSNL/MTNL paid the auction prices, they did not have to acquire licences through the auction processes, they were given the spectrum.) This resulted in operators not being able to leverage economies of scale and reduce prices. Further, the amount of spectrum per operator was limited to 2x5 MHz, whereas international average was much higher. Low bandwidth availability created a vicious cycle as data usage is bandwidth-intensive and low amounts of available bandwidth creates congestion, and hence disincentives for the operator to encourage high data usage. This further impedes lowering of prices.

So is there a possibility of increasing this available bandwidth Since the allocations for commercial spectrum must be made from the existing bandwidth available, we need to examine the existing and planned uses in specific bands.

The Telecom Regulatory Authority of India (Trai) and DoT have worked out plans for refarming of spectrum and for operators in the 800MHz (CDMA) band to go to 1900MHz and for 900MHz (GSM) band to shift to 1800MHz band, as these lower-frequency bands have better propagation characteristics and with technological developments, these bands are more cost effective for 3G and 4G services. Due to the existing licensing conditions, which bundled the spectrum and service licences, the operators holding licences in the lower frequency bands may only provide 2G services and may have to relinquish their existing licences and bid for licences afresh at the end of their licence period. In the new regime, there will be separate spectrum and service licences and winning operators may choose the technology such as 3G or 4G they wish to provide.

While the 900MHz licences are due for renewal beginning 2013-14, by the time pan-India 3G services in 900MHz are offered, it may be at least a few years more. In the 1800MHz band, allocations as per the new auction design and for refarming the 900MHz operators will take away almost all the spectrum available with DoT in this band. The CDMA licences are due for renewal 7-10 years from now. Additionally, since other than one CDMA operator, other CDMA operators have acquired GSM licences also, it is not certain how much they would focus on CDMA in the future. For example, there was only one CDMA bidder in the 800MHz auctions conducted recently. So the value of keeping the 1900MHz free for the potential refarming may lead to a situation where there is little interest in the same.

One simple way of making more spectrum for commercial use available in the 2100MHz band is by DoT exchanging the existing 2x7.5 MHz, currently reserved for refarming in the 1900MHz, with spectrum in the 2100MHz band with MoD, thus making more spectrum available for 3G. Otherwise, this spectrum in the 1900MHz band would lie fallow, at least until the proposed refarming 7-10 years later for CDMA, leading to non-optimal usage of a key national resource. Given the existing spectrum holding pattern between DoT and MoD, this would give both of them contiguous spectrum, allowing for more efficient use.

Also, within 2-3 years, MoD would have shifted to an optical fibre backbone, thus enabling further release in the 1900MHz band for commercial use. This reallocation of spectrum bands between MoD and DoT not only creates efficiency as stated above but also a strategic opportunity to either introduce three more players (each with 2x5 MHz spectrum) for 3G or allocate more spectrum to existing players, if DoT can provide for downlink frequencies in the 2100MHz band from its spectrum pool. Both options would lead to better outcomes, as either operators would be able to provide better quality services or to reduce costs due to competitive pressures.

Thus, the operational task of exchanging spectrum between MoD and DoT could lead to more strategic outcomes. This would require a more consultative approach between MoD, DoT and Trai and between DoT and the service providers. Also, with the commercial value of spectrum increasing and more number of bands likely to have commercial value, DoT would need to have a mindset that is more consultative and can take the initiative to develop linkages with institutions and organisations in the sector and large spectrum users (MoD, police, home ministry) to work out the best solutions for the citizens.

Rekha Jain, Executive Chair, IIMA-IDEA Telecom Centre of Excellence, Indian Institute of Management, Ahmedabad