Skimming the cream

The ongoing developments in the telecommunications sector in India are not oriented to the growth of rural telephony and teledensity, which were the objectives set down in policy statements in the past.

THE telecom sector is once again in a state of flux, with international long distance (ILD) and national long distance (NLD) services being opened for competition. Simultaneously, Videsh Sanchar Nigam Ltd. (VSNL) is slated for privatisation; as a prelude to this, Rs.4,000 crores of its cash reserves has been diverted. Competition has seen an immediate 50 per cent lowering by Bharti Telesonic, a part of the Bharti group, of rates for subscriber trunk dialling (STD) services, leading to even bigger cuts by the government-owned Bharat Sanchar Nigam Ltd. (BSNL). This trend is likely to continue with the opening of ILD. But even as all these developments take place, hardly anyone is talking about rural telephony and teledensity - the supposed objectives of telecom reforms. The issue, however, is a crucial one: with competition, the surplus funds of BSNL, garnered from its long distance revenue and which were used earlier to expand the telecom network, will now be drastically reduced.

The telecom reforms, including the introduction of the element of competition, have little to do with increasing telecom penetration, a crying need today. Instead, the effort is to let the telecom surplus be re-distributed amongst the new private sector entrants and the better-off consumers. The private sector entrants in basic services have consistently refused to honour rural commitments in their licence agreements. In the few States where they are operating today, they are targeting only the top 10 per cent of the consumers. The rest will have to contend with higher local call rates, an increasingly impoverished BSNL, privatised or otherwise, and poorer quality of services. While some sections of subscribers will benefit from lowered long distance costs, the majority will face a bleak situation.

Both the National Telecom Policy 1994 and the New Telecom Policy 1999 identified the need for increased telecom penetration including in the rural areas as the most important driving force of the reform process. The private sector was supposed to bring in the additional capital required to increase the installed base of subscribers, and to take telephones to the rural areas. But it is the erstwhile government monopoly, the Department of Telecommunications (DoT), and its successor, the public sector BSNL, that has fulfilled both these functions. The private sector secured six basic service licences and provided scarcely four lakh telephones in three circles, and the other three are still to take off. In the same period, the much-reviled DoT (its successor BSNL along with MTNL, or Mahanagar Telecom Nigam Ltd.), has provided almost 11 million telephone connections - 4.98 million and 5.93 million in 1999-2000 and 2000-01 respectively. This process, which constitutes an annual growth of more than 22 per cent, has brought their installed base to 32.44 million telephones. The private operators for basic services, the competitors to BSNL, have not provided rural telephones, have refused to pay the committed licence fees, and defaulted on their roll-out targets. They are now asking DoT to pressure BSNL to fulfil all their rural and other commitments, a course a business-friendly government led by the Bharatiya Janata Party is only too willing to take. The private cellular operators have an installed base of about four million telephones and have done relatively better than the basic service operators.

Why did the basic service operators perform so poorly in terms of penetration? In order to understand this, it is necessary to understand the economics of the local network. The licence of the basic service operator is for a circle, roughly co-terminus with a State, and consists of the local and the intra-circle long distance calling network within it. In any such local network today, about 10 to 20 per cent of the subscribers generate the surplus for the operator, about 50 per cent produce a net deficit in revenue and the rest probably break even. If the 10 to 20 per cent segment can be weaned away from the incumbent operators, BSNL and MTNL, the "competitor" will take away all the cream and leave the rest of the network running at a loss. This is why the private operators are targeting only well-off subscribers, leaving the loss-making portion to BSNL and MTNL - a classic case of skimming the cream.

The opening of NLD and ILD was in the offing for some time. The earlier Telecom Regulatory Authority of India (TRAI) led by Justice S.S. Sodhi, had recommended unlimited competition; any party that met the entry criteria and wanted a licence was to be given one. It had also included intra-circle long distance as a part of the national long distance licence. The present TRAI has suggested an upper limit of four NLD operators. The NLD operators are to obtain their licences on paying Rs.100 crores as entry fee and producing bank guarantees for Rs.400 crores. Bharti has already started its NLD operations and has tied up with the cellular operators to carry long distance calls from their networks. The others in the fray for NLD licences are VSNL, MTNL and the Reliance Group. VSNL was promised free entry into NLD operations as compensation for its loss of monopoly in ILD operations.

The recommendation of the earlier TRAI was that the NLD licence also cover intra-circle long distance calls. Fortunately, the current TRAI has recognised the fact that it is the basic service operator who provides the local telecom access on which the rest of the system - STD, ISD, Internet and other services - rides. The intra-circle long distance revenues account for about Rs.8,000 crores out of the Rs.15,000 crores of total revenue from domestic long distance traffic. If intra-circle long distance is opened to competition, this will either make the local services unviable or lead to a steep hike in local call charges and rentals.

The licence conditions for ILD are even more relaxed than those for NLD operators. There is no cap on the number of operators and any prospective player can secure a licence after paying a fee of Rs.25 crores. ILD operators will have to pay a licence fee of 15 per cent of their revenue. The TRAI has also permitted voice over the internet protocol (VOIP) as a lower quality and cost service that subscribers may avail as an option. VOIP services do not mean Internet telephony for the subscriber; it only means that an ILD operator may deploy Internet based routing and transmission of traffic while connections at both ends are through normal telephones. In Internet telephony, a subscriber has the option of using a computer at his or her end to connect to another subscriber using an ordinary telephone. The two services will be distinct, and presumably there will be a large price differential between the two to compensate for the loss of quality that VOIP will involve. The roll-out commitments are also not significant; an ILD player will have to invest about Rs.275 crores for four gateway switches over the next three years. As an observer has commented, this will allow even fly-by-night operators to enter the sensitive ILD market.

The net result of freeing the ILD market will be a free-for-all in the estimated Rs.8,000-crores-worth ILD market, and a huge drop in VSNL's share value. Of course, the surplus from this segment of the market, that now supports the local network expansion, will also dwindle as it is now supposed to be a part of the licence fee of 15 per cent. Interestingly enough, while there is no bar on any private player offering ILD services, the government is thinking of barring MTNL and BSNL from entering this sector for two years.

Apparently, competition is good only if it comes from private players; or is this an implicit admission that if there is an even playing field, private players cannot compete with public sector undertakings?

THE major policy focus in the telecom sector should be the improvement of telecom penetration. Currently, it is only 3.16 telephones per hundred (March 2001), well below the world average figure of 10 per hundred. Even though the teledensity is poor, India is today the eighth largest telecom network in the world.

There is no doubt that increased competition will significantly lower long distance charges. But the question that needs to be asked is: as the current network expansion of more than 22 per cent a year takes place from this surplus, how long can the local network be expanded at this pace? Will local call charges and rentals have to be raised in order to augment the revenues of the local network operator so that the pace of increase in teledensity continues?

The intimate connection between local call rates, rentals and long distance charges became clear from the first re-balancing that the TRAI carried out in 1999. The principle adopted for telecom expansion in all countries in the initial phase is to make access charges low and cross-subsidise local charges from long distance revenues. However, the TRAI felt that it should switch to "cost-based" tariffs and remove such cross subsidies. This resulted in access charges and local charges doubling under the revised tariffs, while long distance charges were lowered. This also reduced the surplus available with BSNL for expansion. In the first year after re-balancing, the surplus (of BSNL and MTNL) dropped to Rs.8,500 crores from Rs.10,000 crores the previous year. There is no question that with the sharp drop in long distance rates, there will be a further erosion of BSNL's revenue. Some have argued that the universal service obligation (USO) levy, raised through a share of the operator's revenue, can meet this shortfall. A back-of-the-envelope calculation will show that the decline of long distance rates cannot be offset by the USO levy.

WHY bother about increased penetration at all? Why not provide telephones only to those who can pay its "true" costs? The problem with this approach is that telecom is not just another commodity; it is not a brand of soap or detergent. Infrastructure facilities, such as power and communications, are prerequisites for economic development. If these do not exist, the economically backward areas will not be able to develop, leading to intolerable strains on the nation. Can anyone doubt the fact that the lack of economic opportunities in the northeastern region have provided a fertile ground for militancy? Or that in Srinagar, popular alienation is exacerbated by the fact that often only two hours a day of electricity supply is available during winter?

Cheap telecom access acts as a motor for development. If telecom activity is viewed as a purely commercial one, there can be no economic justification for providing telephones to either rural or remote areas; or for keeping the cost of connecting to the network - rentals and installation charges - low. Only the larger social and economic picture explains the need to expand the sector.

All countries that have reached a high level of telecom penetration have done so by initially cross-subsiding local access charges from long distance calls. This is true even of countries where telecom has been solely in private hands. AT&T in the U.S. subsidised its local network from its long distance revenue. It is only after virtually every home had been connected that the U.S. introduced competition in long distance and re-balanced tariffs.

For increased teledensity, the cost that a subscriber has to pay for connecting to the network - the basic access charge - must be kept low. If the access charge is high, demand for telephone connections will fall and teledensity cannot increase. High long distance rates have helped keep the access charges low while generating funds for the expansion of the network. The need for higher long distance tariffs has little to do with its actual cost. The direct cost of the long distance portion is only about 15 per cent of the total cost of connecting the subscriber to the network. If we take the cost of Rs.27,000 for installing a phone today, the major part of it - about Rs. 23,000 - goes towards connecting the subscriber to the local exchange.

The pace of penetration has improved markedly in recent years, and considerable surplus was available from the long distance revenue. At current rates, BSNL and MTNL are slated to double the number of telephone connections in the country every five years. If the current surplus were maintained, the teledensity would have increased to 7.5 in the next five years and 15 in the next 10 years. With the fall in costs per line and cellular operators supplementing this effort, a teledensity of 7.5 to 10 by 2005-06 is a realistic target. The current assault on the long distance surplus endangers this target.

The key to increased penetration is the local network operator, or, to use the Indian term, the basic service operator. It is the viability of the basic service operator that determines the scale of telecom penetration. Fortunately, opening NLD to competition still allows basic service operators to retain intra-circle long distance revenues, unlike the recommendations of the earlier TRAI. However, as the surplus from inter-circle long distance drops, some rate change in local call rates is bound to follow. The TRAI will have to work on any tariff re-balancing carefully, so that high end-consumers pay higher call charges and low-end consumers are allowed to connect to the network at low costs. This means that in any re-balancing of tariffs, rentals and installation costs should not be increased. Instead, innovative methods such as increased peak hour call rates can be pursued to increase the revenue of the local operators. Otherwise, there would be an immediate slow-down in telecom penetration.

Prabir Purkayastha is Secretary, Delhi Science Forum.