The Hindu Business Line : For a smooth transition to 3G

In a recent interaction with the press, the former IT and Communications Minister, Dayanidhi Maran, had indicated the general principles based on which the 3G Policy details could possibly be worked out.

Let us take a look at the key features of the approach, and their implications for the telecom sector and consumer.

It has been stated that 3G is a new service and not a continuum of existing 2G services. It has further been reported that there is an availability of 42.5 MHz spectrum for these services. Out of this spectrum, possibly 40 MHz is available in the so-called 3G band i.e., the 2.1 GHz band.

The third major aspect is that 2x10 MHz blocks are being considered so that there will be four 3G operators. However, one slot is being reserved for BSNL/MTNL so that there can be a maximum of three private operators. Last, the operators will be chosen on the basis of an auction in which foreign investors will also be able to take part.

Ensuring fair access

As the name suggests, 3G is the third generation of the mobile technologies and was created to be able to provide additional services that could not be provided by 2G technologies. The technology specifications were drawn up in a manner that ensured backward compatibility with 2G services to ensure that the investments made in 2G infrastructure do not become infructuous with the advent of 3G.

There is, therefore, a direct link between 2G and the 3G technologies. It can be argued that this does not necessarily imply that 3G should be a continuum of 2G in the sense of automatic allocation of free spectrum to 2G players.

However, the fact remains that 3G is the evolution path of 2G and every operator who is providing 2G services has, in a way, a right to grow further. Therefore, if the government decides that it will allocate 3G spectrum at a price — a decision which is entirely within the domain of policy making — it certainly must ensure that all existing players have access to at least enough spectrum in 3G band so as to have a chance to compete and not become sitting ducks in the intensely competitive mobile services arena.

The approach of allocating 10 MHz to each 3G operator would apparently have been on the premise of greater efficiency in spectrum utilisation. However, with only 40 MHz of the possible 60 MHz spectrum in 2.1 GHz band being available at this point of time, this approach will restrict the number of 3G players to only four. Out of these four slots, one slot is proposed for BSNL/MTNL and therefore only three private operators can get 3G spectrum. With the approach of permitting new players, especially foreign operators, to join in the 3G spectrum race (a perfectly reasonable policy), it may so happen that less than three (theoretically none) of the existing six or seven operators may get a chance to grow. Besides, the competition for a slot of spectrum will be so intense that the spectrum costs will sky rocket (no matter how well the auction is designed), resulting in a direct impact on cost of 3G services. Also, this does not augur well for competition in the 3G market. The low tariffs in the 2G services in India have come about due considerably to intense competition. Unduly high competition at the bidding stage will have the opposite effect.

The solution to the problem lies in reducing the granularity in spectrum lots i.e. allowing 5 MHz lots at this stage so that a total of eight slots are available. Like any other solution this one also has pros and cons. The obvious downside is the aspect of thin spread of the spectrum, especially if one looks at the UMTS Forum suggestion that a minimum of 20 MHz per operator should be made available (15 MHz for FDD operation and 5 MHz for TDD operation). Obviously so much spectrum is not on the horizon in India. One possible solution is that when the remainder 20 MHz spectrum in this band becomes available, further 5 MHz slots can be auctioned to the then existing 3G operators to permit them increased availability of spectrum. An announcement of such a policy at this stage will take care of the concerns of inefficiency of spectrum utilisation besides permitting the industry to follow the merger and acquisition path of stabilisation in an atmosphere free of duress - a scenario which will evolve if the current proposed approach of limiting the number of players by regulatory action is followed. The other possibility is to follow the Australian example where a combination of 5MHz and 10 MHz was auctioned. With four 5MHz slots and two 10 MHz lots, a total of six operators can come in and if, as discussed below, 5MHz in the 450 MHz is also made available, a total of seven operators can be accommodated. This, with the possibility of permitting a company to bid for up to two 5 MHz slots, is a very attractive option.

Growth path for CDMA

Another aspect requiring consideration is the growth path for the current CDMA operators. While it is true that they can also bid for the 2.1 GHz slots (and they probably will), the fact remains that in the interest of the industry and the consumers, bands where economies of scales due to worldwide deployments are available should also be exploited. It is further true that the government has permitted trials of EVDO equipment of CDMA in the 1900 MHz bands; even after the trials, the decision to go for mixed band allocation (2.1 GHz and 1900 MHz) at the same location will be a long drawn one and a difficult one. On the other hand, at least for rural operations, 450 MHZ band can be utilised where EVDO equipment is readily available. In addition to the above, the 3G Policy must also come out with a clear permission to the existing operators to offer 3G services in their existing 800 and 900 MHz bands. This is of advantage to the consumers since it is well known that 3G equipment is more efficient than 2G equipment for handling voice traffic, as a result a greater number of voice-only subscribers can be catered to in the existing allocations.

Incidentally, the Telecom Regulatory Authority of India (TRAI), in its 2006 recommendations, has, by and large, stated what has been concluded above and therefore it will be best for the policy makers to accept those recommendations in toto rather than modify the spectrum lot size, especially since the indicated spectrum availability in 2.1 GHz band is better than what was presumed by TRAI when it released its recommendations. Implementation of TRAI recommendations, augmented by a modified Australian model, will go a long way in meeting the expectations and the aspirations of all stakeholders.

(The author is a former chairman of BSNL and former Member, TRAI.)