Licensing and spectrum remain critical challenges
While the government goes about trying to create a digital India, as a recent Trai consultation paper on broadband points out, the country remains woefully behind what is needed. There are several reasons for this, and Trai talks of the poor progress of the national optical fibre network which is being developed by PSUs and not private firms. There is also the issue of how BSNL has not been able to convert enough of its existing land-line customers into broadband ones, but the real reason lies elsewhere. It lies in the fact that the government simply hasn’t done enough to vacate adequate spectrum for mobile broadband; indeed, unless some of its policies are quickly reversed, by the government or through court action, broadband access is going to become very expensive and will naturally slow down. While details vary for each spectrum band, what’s common to all is that, with the defence ministry not releasing spectrum, optimum use is not being made of even the spectrum that firms have. In the 1800 MHz band, for instance, firms have been given spectrum in 200 kHz chunks which make them unsuitable for using to meet the data needs of customers—over two-thirds of the spectrum is not contiguous, rendering it useless for data usage. In the 2100 MHz band, ideally suited and used today for mobile broadband, just 20 MHz—a third of the total available—has been released for commercial use while another 15 MHz of spectrum of 3G spectrum is being wasted as this has been reserved for CDMA firms which cannot use this as the existing CDMA 800 MHz devices do not work in the 1900 MHz band, unlike GSM ones which work on both 900 MHz and 1800 MHz.
The problem gets worse in the 900 MHz band that can also be used for more data services. Though a bit better than the 1800 MHz band, just half the spectrum here is contiguous and suited for data usage. What complicates matters is the government’s refusal to renew licences in this spectrum band after their initial 20-year period was over. As a result, in the February 2015 auctions, firms that will be bidding for spectrum will be fighting a do-or-die battle—if they don’t win back their 900 MHz spectrum, they will have to shut shop; the matter is in court, so a court-based solution is a possibility. If spectrum costs skyrocket, as they will, this will slow the progress of broadband in the 900 MHz band. An alternative, a few years down the line, is 700 MHz spectrum, but since the 900 MHz auction will become the base price, this will also be prohibitive and will prevent telcos from buying the large chunks—typically 2x10 MHz—of spectrum they need to drive better broadband speeds.
This is the issue the government needs to grapple with, but is not showing too many signs of being able to do so. Unless adequate amounts of spectrum from mainstream spectrum bands—900, 1800 and 2100 MHz—are brought into the next auctions, bid valuations are going to keep shooting up. Theoretically, with less firms left in the fray, the big boys may still manage to stay afloat. But they will do so only after hiking costs of broadband usage. While there may still be enough of a market at even higher prices—which works fine for the firms who are interested in making profits—this will prevent the spread of broadband to the extent India requires. And, as the Trai consultation paper points out, a 10% increase in broadband penetration accelerates economic growth by 1.4%.