Adverse Impact on BSNL

As those that do not wish to avail of the OneIndia facility would stay with the current rentals and free calls, the drop in long distance revenue would have to be made up by greater call volumes that might be generated by the lower call rate. It is unlikely that such a high surge in volume would take place. If however, such a surge were generated, BSNL would also have to make considerable investments for upgrading the long distance network. Either way, slashing long distance tariffs by a whopping 58 per cent can only adversely affect BSNL.

In order to reduce the long distance calls to Re 1 a minute, there was also a need to bring down the ADC, which used to be a substantial portion of every long distance call charge. Earlier, 30 paise for every call was an ADC levy, which resulted in an annual transfer of about Rs 5,000 crore to BSNL for subsidising its rural operations. TRAI has finally changed the ADC regime from a call-based one to revenue-sharing and reduced this transfer by about Rs 1,800 crore. It has provided for a revenue share of only 1.5 per cent of the total revenue of all operators to go to the ADC account and has also reduced international ADC rates by more than 50 per cent (to Rs 1.60 for incoming and Re 0.80 for outgoing calls).

The net result of all this is that BSNL and Mahanagar Telephone Nigam Ltd (MTNL) are likely to lose Rs 3,000-4,000 crore of their long distance revenue, even

Economic and Political Weekly March 11, 2006 after higher landline rentals are taken into account. With the additional loss of Rs 1,800 crore from the lower ADC levy, Maran has at one stroke converted what were still thriving PSUs, even under a strong competitive regime, to possible basket cases. Effectively, BSNL is being asked to take a major hit in its revenue, while companies that are flouting the terms of their licence of providing 10 per cent rural telephones get away scot-free.

It is not that Maran’s idea of long distance tariff being based on a flat rate rather than distance is without any merit. There is little doubt that the model of telephony is changing from the earlier distance-based circuit-switching network (point switching telecom network or PSTN) to the current data-based packet switching networks. The internet model and the earlier voice-based telephone models are quite distinct. In the earlier predominantly voice networks, distance was the basis for charges: the longer the distance, the more you had to pay. However, the data networks do not care what the distance travelled will be: the internet does not bother about either the route or the distance that the data packets travel. As internet and broadband increasingly dominate the telecom scenario, with even voice calls being transmitted over the internet, the internet model with its death of distance will emerge as the dominant revenue model.

The issue here is not that revenue models have to change from a distance-based one to a data-based model, but the timing and extent of this change. In all telecom networks, the long distance revenue has always been used to keep the cost of telecom access low. The simple argument for doing so is that unless the telecom network expands, call volumes and, consequently, revenue cannot grow. Only the rich calling each other does not generate enough traffic; they also need to call a whole range of other people users who may not be willing to pay for high cost connections. So if we want increased telecom penetration and high teledensity, we need to keep the costs of connecting to the network low. Once we have achieved a high teledensity, we can then afford to raise the costs of telecom access. But if it is done before this, it is likely to deter a number of people from installing telephones. So instituting a telecom regime where the long distance calls do not provide any surplus for the local network can only force the local call rates and access charges to rise and work against the immediate goal of increased teledensity.

It is not that the long distance call rates in India are high. They have come down dramatically in the last few years and are well below rates in most countries. It costs less for a mobile subscriber in Germany to call India than to make a local call there. Therefore, the argument about needing to lower long distance rates here and now makes very little sense. It is true that as we go forward with voice over internet protocol (VOIP), etc, a model free of distance would need to be implemented. The question here is when this should be done, and, after making the tariff distance free, what it actually should be? Maran’s decision to impose a one-rupee tariff without addressing how to pay for the low cost access required to increase teledensity, could cost the country dear. And if it also means bankrupting BSNL, this would be a double disaster.

If we look at the telecom scenario currently, we see that the telecom boom has not only passed the rural subscribers by, it has also sharply widened the ruralurban gap. As TRAI itself has noted “the large differential between rural (1.94 per cent) and urban teledensity (31.1 per cent) cannot be sustainable. The Authority recognises that without focus on rural areas, sizeable growth in telecom sector would not be possible” (TRAI press release, October 3, 2005). The second disquieting trend in telecom is the virtual stagnation of landlines, while the mobile sector is still maintaining its rapid growth. Of the 32 million new connections between April 2005 and January 2006, 31 million were mobile phones. Obviously, the structure of voice telephony market is shifting rapidly.

BSNL’s Role in Rural Telephony

TRAI has also noted in its consultation paper on interconnection usage charge review (March 17, 2005) that the installation of village public telephones (VPTs) has come down from around 60,000 in 2001-02 to about 15,000 in 2004. The rural lines added in the same period have also dropped by about 40 per cent from 2.3 million in 2001-02. In the same period the mobile “revolution” has passed over the rural areas and the smaller towns. As Table 1 shows, the rural areas have virtually no coverage and cellular networks cover only about 1,700 towns out of 5,200 towns in the country and have a population coverage of 20 per cent.

Let us look at the rural telecom growth patterns. It is obvious that rural telephones are more costly to install and also generate lower revenue. Unless there are either penal provisions or incentives, telecom service providers would not readily install rural telephones. This can be seen from Table 2, which shows that almost all rural telephones currently are being provided by BSNL.

The figures are quite astounding. Not only have the basic service operators not

Towns ~1700 out of 5200 ~200 Million Rural area Negligible Negligible

Source: ‘Interconnection Usage Charge Review’, TRAI Consultation Paper No 4/2005, March 17, 2005.

Table 2: Service Providers and Percentage of Fixed Wireless and Rural Lines

Service Provider Name of the Circle/ Per cent of Fixed Wireless Per Cent of Rural Lines Service Area Lines in Operator’s in Operator’s Fixed Fixed Lines Subscriber Lines

BSNL All India (except Delhi and Mumbai) 2.60 35.20 MTNL Delhi and Mumbai 1.09 0.00 Bharti Delhi, Madhya Pradesh, ,

Haryana, Chennai 3.46 TATA Maharashtra, Mumbai, Andhra Pradesh,

Delhi, Gujarat 77.39 0.23 Shyam Rajasthan 18.49 3.37 HFCL Punjab 24.53 0.45 Reliance All circles except Assam

and North-East 97.27 0.66 Total 7.70 28.93

Economic and Political Weekly March 11, 2006

provided rural telephones, bigger the player, the smaller the number of rural telephone lines they have provided. Major players like Reliance and Tatas have not even provided fixed lines. They have preferred to use the wireless route in order to keep their capital costs low and attack the high-end market. As against less than 1 per cent of total fixed lines as rural telephones being provided by private basic operators, fully 35 per cent of state-owned BSNL lines are in rural areas. Obviously, without BSNL, we would not have any rural phones in the country.

The private players have willingly paid a penalty, pegged at a mere rap on the knuckles than a serious one, rather than fulfil their licence terms and conditions. Both TRAI and the Department of Telecom have preferred to wink at this continued violation by private players while bemoaning poor rural teledensity.

TRAI has now suggested creating a shared infrastructure out of the Universal Service Obligation (USO) funds available for all players to increase rural teledensity. While this measure may indeed help in increasing teledensity, why cannot far more stringent penal measures be imposed on those who violate their licence terms and conditions is a question that TRAI still needs to answer.

However, the issue here is how does Maran’s OneIndia scheme and lowering of ADC levy help or harm the rural-urban divide. Here, we need to address the second part of the OneIndia scheme: the tariff of Re 1.00 a minute could be implemented only by removing the ADC from long distance calls. This was why the ministry was arguing that instead of ADC being call based, it should be realised through revenue sharing. Even TRAI was moving in this direction, as it had announced a merger of ADC and USO levy by 2008-09.

Objective of USO

The genesis of USO and ADC are quite different. The Universal Service Obligation or USO stems from the need to provide telecom access to all areas in the country irrespective of their geographical location. As we still have about 85,000 villages without a telephone connection, USO is geared to primarily extend the telephone infrastructure to these areas. In this sense, it could be viewed as a one-time investment for providing telecom infrastructure in areas where such infrastructure does not exist today. The access deficit charge or ADC addresses quite a different issue. It deals with the need to keep the costs of connecting to the local network low. It is obvious that what a person will pay for a telephone depends on this cost as a proportion of his or her income. In a country, where the average income is still abysmal, it is imperative that the cost of access to the telecom network should be kept low. Earlier, it was accepted that the true costs of the local network should not be recovered from local calls and rentals alone and therefore long distance revenue was used to subsidise the local network. As we have outlined earlier, it is obvious that this subsidy from long distance revenue is not going to continue for too long. TRAI’s solution has been to raise the local access costs (re-balancing the tariffs) and use a percentage of revenue of all telecom providers including mobile and long distance operators to provide the necessary transfer to the local network.

This brings us to the next issue, who should get the ADC funds that are generated from either a call-based or a revenueshare regime? Clearly, the rural subscribers not only have higher costs, they also generate less revenue. So those with rural subscribers should get the ADC levy. But this is not all. There are about 40 per cent of subscribers in urban areas that generate less revenue than the cost of providing them with services. Obviously, a service provider that caters to such low-end consumers also deserves a part of the ADC funds. If we look at Table 2, we will see that almost the entire rural subscriber base is with BSNL. The other column – percentage of fixed wireless lines in operator’s fixed lines – is also revealing: it shows that most private fixed line service providers are installing fixed wireless lines which have higher call rates and not fixed landlines. So if access deficit is to be compensated to the fixed line operators, those offering a bulk of their lines using the wireless route should not get a share of the ADC. Clearly, only those who are providing fixed landlines should be given these funds.

The decision of TRAI that other unified and fixed line operators should retain the ADC amount, including that from outgoing international calls, is quite surprising. If ADC is for covering the cost of lowrevenue-generating rural subscribers and low-end urban subscribers, there is no justification for the ADC amount to be retained by those who are not serving this market. This differentiation, where only the mobile licensees have to pay a percentage of their revenue but not operators who provide fixed wireless, does not seem justified on any count, particularly if we take into account that the fixed line operators are the ones violating their licence terms and conditions by not providing rural telephony.

Were other methods available to TRAI to recover local network costs? We have argued earlier that one simple method, which is very widely followed internationally, is to change the pulse rate for peak and non-peak hours so that local network costs can be recovered fully. This would be a simple matter as the pulse rate at the exchange level can be changed very easily based on the time. This would ensure that calling rates for the low-end consumers do not go up while ensuing a larger revenue for the local network. The expansion of the bandwidth of the network is directly related to the peak traffic on the network and requires major capital investments. It is therefore quite fair that those generating peak traffic should pay more for their calls. Strangely enough, neither TRAI nor the PSUs have shown any interest in this method of recovering local network costs.

While the issue of a flat rate and ADC are important as they pertain to the current evolution of the network, there are larger structural changes taking place that also needs to be addressed. It is now obvious that voice traffic will increasingly shift to the mobile networks. So how do the PSUs with their huge legacy investments in copper cables compete in the market? It is time the PSUs realise that what they view as their sunk costs – the copper cable network – is precisely their strength. They need to think that in the future, data networks will largely be land-based while the voice network will be wireless. The focus for their landline operations has to therefore shift to the internet and internet-based services. Here again, MTNL and BSNL are concentrating on only numbers and leaving the high value segment to be creamed by the private players. Unless the two can develop a business model that takes into account the emerging structure of the telecom market, a mere defensive battle is not going to be enough. BSNL and MTNL have to learn that business as usual does not work when the usual business is changing. Email: prabirp@gmail.com

m

Economic and Political Weekly March 11, 2006