At 1.4 per cent broadband density, India is sitting at near bottom of the ladder worldwide. The average broadband connection speed of 1 Mbps is the lowest in Asia-Pacific region. This is the status after nearly a decade of launching the Broadband Policy in 2004.

Growth prospects

Fixed lines have made launching of broadband services possible but further growth prospect is limited due to the lack of development in fixed access infrastructure. Some further contribution can be expected from TV cable-based offerings. However, for a real kick-off, similar to the growth of 2G services, mobile broadband is inevitable.

This trend is also confirmed by the fact that by mid-2012 the numbers of hits for accessing Web pages on mobile handsets have already surpassed those of PCs and laptops. However, for mass adoption of mobile broadband services right-priced smartphones are a must, since high-cost smart customer terminals constitute a major entry hurdle.

The Indian market is characterised by 91 per cent handsets in use which cost less than Rs 5,000. Thus, smartphones which cost less than Rs 5,000 and preferably less than Rs 2,500 are a must as a terminal equipment for the popularisation of broadband. This requires a large market and the Indian market can be considered large only if the price segments mentioned above are targeted. This can only happen if manufacturers are able to make use of global economies of scale i.e. services are provided in India in globally harmonised spectrum bands as was done for 2G services.

Mass adoption

So far we have looked at the entry cost for a consumer. The 2G growth example shows that mass adoption also requires the cost of service to reach levels of mass acceptability.  For broadband services, this means affordable bandwidth costs since apps that are becoming popular in India will require higher bandwidths. To arrive at this affordable cost level, it is essential that competition becomes more intense with each individual operator having exclusive spectrum slots capable of offering mobile broadband services.

Mobile broadband service can be provided through either 3G (HSPA) or 4G services (LTE). While 4G spectrum was auctioned in mid-2010, most of the prospective service operators have not launched the service and whatever token service has been launched has not yet attracted a large user base.

This is not surprising since neither low-cost handset-type LTE instruments are available nor does the present license permit voice (as VoIP) which for India continues to be the main factor at the moment for business case. Thus, in the near future any major contribution by 4G services to help the popularisation of broadband services is unlikely. 

3G services, launched a few years ago , are at best a modest success. The band chosen for 3G services is indeed the right one since 2.1 GHz spectrum enjoys global harmonisation and global economies of scale.

However, with the present thinking and approach, the possibility of growth through more spectrum allotment is ruled out. Further, bandwidth costs to the consumer are high with insignificant competition and no operator (other than the combined BSNL and MTNL offering) offering pan-India coverage.

The symbiotic spectrum sharing arrangement between three private operators to create virtual pan-India footprint, is under regulatory and legal dispute. These factors are bound to discourage operators from pursuing heavy growth and making significant financial commitments.


The only way in which 3G and therefore broadband scenario can be brought back to one of rapid growth and expansion is to introduce genuine competition and more spectrum than the present 5 MHz slots. Initially, the government seemed to be banking on 2.3 GHz and 700 MHz bands for broadband growth through LTE. However, as discussed above LTE can at best contribute significantly only in the medium term. Refarming of 900 MHz band is being considered as a solution.

Quite like the 2.1 GHz band, 900 MHz band is an excellent band for India because of its high global harmonization. However, the problem in using the 900 MHz band appears to be the staggered vacation (starting next year) that stretches up to 2020 or so.

Further, the question of legacy 2G Network will remain. In case existing operators in this band again succeed in winning spectrum, they may like to retain a major part of the spectrum for their existing remunerative 2G network giving lesser weightage to new 3G network. It will certainly neither be possible nor desirable for the licensor to insist on using the band for only 3G purposes.


Recently a solution has been proposed for creating 3 more 5 MHz slots in the 2.1 GHz band. As per the present agreement between DOT and DOD, 150 MHz of the total 300 MHz from 1700 to 2000 MHz has been allotted to DOT. This includes 25 MHz for 3G applications with 20 MHz already auctioned (including that given to BSNL or MTNL). A further 15 MHz (1900-1907.5 MHz plus 1980-1987.5 MHz – actually 1900-1910 and 1980-1990MHz) has been reserved for 3G EVDO services.

Disregarding, for the moment, the dispute whether such a deployment will interfere with the adjacent GSM band 1954-1979 MHz, let us dispassionately look at the existing situation.

The last auction for CDMA services in 800 MHz band did not get any bidders and fresh auction is expected to interest one or two operators only. Such requirements of the next two or three years can be met from 800 MHz band.

If an arrangement is made with DOD to barter the 1900-1907.5 and 1980-1987.5 MHz for 1939-1954 MHz (15 MHz) band on immediate basis, it is possible to create 3 more slots for 3G since corresponding band 2129-2144 MHz is already lying unused.

When the Optic Fibre Network for the defence forces is ready, negotiations (as per agreement) can be held to get back the 1900-1907.5 and 1980-1987.5 MHz for EVDO. Such a solution seems entirely workable without jeopardizing CDMA interest.

(The writer is former Member, TRAI and former Chairman, BSNL)

(This article was published on February 7, 2013)
XThese are links to The Hindu Business Line suggested by Outbrain, which may or may not be relevant to the other content on this page. You can read Outbrain's privacy and cookie policy here.