SC right in striking down call-drop order
The Supreme Court has done well to strike down the Trai penalty on call drops since, as this newspaper has been arguing since the order was put out, it never made any sense —ironically, much of this was admitted by Trai itself in a technical paper issued a month after the initial call-drop one. Apart from the fact that no commercial cellular network can offer 100% coverage with no call drops—but this is what the Trai’s one rupee penalty per call amounted to assuming—Trai did not take into account very serious challenges faced by mobile phone operators. While customers do have a right to good quality service, Trai did not consider that, as part of government policy, spectrum licenses were not automatically renewed after 20 years and telcos had to rebid for this in auctions. Since the spectrum they won back—at large costs which, in turn, reduced their capacity to invest in other infrastructure—was in different frequencies, and sometimes in different bands as well, telcos needed to re-optimise their networks, a process that can take 12-14 months; the technical paper, not surprisingly, offered this as a big reason for the rising call drops.
Similarly, it wasn’t as if the telcos didn’t want to invest in setting up more telecom towers; more often than not, the municipal authorities didn’t allow this. Also, every mobile network has an optimum number of towers—put in too many, and their signals will clash. So, all major cities like Delhi, Shanghai and Singapore have a similar tower density of 2.2 towers per square km, if you leave out areas like Lutyens Delhi where there is a problem in getting permissions. What also makes a big difference is the amount of telecom spectrum a telco has, and the optic fibre between the towers to take the signal to the telco’s switch. If a Bharti Airtel, for instance, had 10MHz of spectrum in the 900MHz and 1800MHz frequency bands instead of the 6MHz and 7MHz that it does (respectively) in Delhi, it would need 4,000-4,200 towers instead of the 6,000 or so it has today. If, on the other hand, it had 5MHz of 900MHz and 15MHz of 1800MHz, it would need around 5,500 towers—if there is more spectrum in a lower frequency, a telco needs less towers.
As a result of the shortage of spectrum, Indian telco networks tend to be a lot more ‘loaded’ than those overseas and carry more voice/data traffic per unit of capacity. A top telco in Delhi will be dealing with 3.8 million minutes of voice traffic per day right now versus 4.6 million in Shanghai and 0.6 million in Singapore. Given the number of sites and spectrum, a Delhi telco will be handling 49 hours per MHz per site—the comparable number is 6.5 for Shanghai and 8.9 for Singapore. For data, this becomes 3.4 GB per MHz per site versus 0.2 for Shanghai and 2 for Singapore. Not surprising then, that calls drop or that the internet crawls. With telecom minister Ravi Shankar Prasad quite focussed on providing more spectrum by way of sharing/trading policies, harmonisation of existing spectrum—this creates more blocks of contiguous spectrum that is more efficient to use, especially for data—and even bringing in more spectrum into the auction from the defence sector, hopefully this will soon be a thing of the past.