Chap 10 2003

Download as pdf or txt
Download as pdf or txt
You are on page 1of 13

A Review of Telecom Regulatory Authority of Indias Tariff and Interconnection Regulation 239



A REVIEW OF TELECOM REGULATORY AUTHORITY OF INDIAS TARIFF AND INTERCONNECTION REGULATION


Rekha Jain

Tariff rebalancing and interconnection regulation plays a critical role in reforming the telecom sector, characterized by rapid technological changes and monopoly service provision by the incumbent. Due to rapid advances in technology, the prices of long distance communication have fallen more dramatically and incomparably more than that of the wire line local access. Further, competition in the long distance segment has resulted in a sharp decline in prices, almost bringing them near costs. However, the decline in long distance prices may not be accompanied by costbased local access rates. They may have been historically kept low due to political reasons. The below cost rentals and call charges may not invite private participation. On the other hand, it may not be possible to have above cost long distance charges (usually used to subsidize the below cost rentals and local call charges) in a competitive environment. Tariff regulation to rebalance these by increasing local call charges as a means to encourage private entry and corresponding reduction in long distance charges creates the appropriate incentives for investment. Tariff rebalancing as a regulatory tool can ensure adequate incentives to the incumbents for the transition to a competitive environment by specifying the road map and protecting its revenues.

interconnection to the new entrant. With the deregulation of the Indian telecommunications sector and the move from monopoly to duopoly and finally to oligopoly in the telecom sector, the choice for the end user has increased. However, this has also placed demands on the regulator to ensure implementation of policies for efficient operation of the sector with interconnection as one of the critical issues in enhancing competition. One of the most crucial and debated issues in interconnection regulation is interconnection charging as it forms a major part of both the expenses as well as the revenues for a telecom operator. Interconnection charging can be a significant factor in the viability of the business of an interconnection-seeking private telecom operator as well as for the interconnection-granting incumbent. The access charges, if passed on to the consumers, also affect the retail price of services offered. Therefore, the charging regime has to be efficient, fair and unambiguous to protect the interests of all the players. TRAI began tariff balancing as early as in 1999, and the exercise has gone through two phases (Box 10.1).

PRIOR EVENTS
In May 1999, the TRAI allowed fixed service operators to provide WLL services using CDMA as it would be strongly in the interest of the subscribers. Technologically, this was a cellular service1. However, TRAI had limited the mobility that could be offered on these networks to the Short Distance Charging Areas (SDCA). The spectrum for these services was given on a first come first serve basis and the players
For a detailed discussion of this initiation and the likely fall out see Annexe Table 10.1, Jain and Sanghi (2002).
1

Interconnection to the Fore


With multiple players in the industry, interconnection has become a critical issue, especially as the incumbent usually has a large spread out network. The challenge before regulators is to provide a framework that addresses not only the technical, economic, and commercial aspects, but also provides for technological changes. The regulator must adopt policies that encourage the emergence of a competitive market, despite the reluctance of the incumbent to provide

240

India Infrastructure Report 2004


Box 10.1 Tariff and Interconnection Regulation (The Earlier Phases)

THE FIRST PHASE


TRAI began the tariff rebalancing with a second consultation paper in September 1998 and specified the tariffs in the first Telecom Tariff Order, March 1999 (TTO 99) (http://www.trai.gov.in/order.html).

Tariff Rebalancing
For fixed services, the TTO 99 was the first step in the process of tariff rebalancing. The tariffs were specified in terms of standard packages that all providers were obliged to offer. However, there was provision for both fixed and cellular service providers to offer alternative packages in terms of monthly rentals and per minute charges. The TTO 99 envisaged an increase in monthly rentals and a decrease in NLD and ILD tariffs to bring them near costs in 3 phases over a 3-year period. The phasing would help to cushion the impact of changes. The TTO 99 categorized the users as rural and urban (based on classification as per census) and further within each category, the users were divided into low, general, and commercial. The rentals for the different categories varied and also depended on the exchange capacity to which the subscriber was connected (Annexe 10.1). The categorization into low and general user was based on the usage in terms of MCUs. For both the urban and rural users, low users were those who made less than or equal to 500 MCUs per month, while general users were those who belonged to neither commercial nor low user categories. The definition for commercial user was left to be decided in the future after a due process of consultation. For the time being, commercial users were those who opted for a rental under the commercial category. For rural subscribers, if MCUs were less than or equal to 500, then the per call unit charge would be Rs 0.80, additional MCUs would have a charge of Rs 1.00 per unit, while for urban areas, the corresponding charges were Rs 1.00 and Rs 1.20, respectively. The per minute charge did not vary within the specific sub-categories (low, general, or commercial). The rural and urban subscribers got 75 and 60 free MCUs per month. The paper provided for a change in rentals for every year for the 3-year period for the period 1 April 2000 to 31 March 2002 only for the general category. For the long distance and international calls, the TTO 99 specified a pulse rate and charge on different distance slabs. These were expected to reduce over the 3-year period. For the rural subscriber the minimum and maximum charges for a one minute long distance call at pulse charge of Rs 0.80 per minute would be Rs 0.80 (for a call within 50 km) and Rs 20.00 (for a call distance of greater than 1000 km). These were expected to be Rs 0.80 and Rs 14.40, respectively by 31 March 2002. For the urban subscriber the corresponding rates would be Rs 1.00 and Rs 25.00, expected to go to Rs 1.00 and Rs 18.00, respectively, over the 3-year period. The above elements were a part of the standard package which all service providers were required to provide. For cellular services, the TTO specified cost based rentals (Rs 600 per month) and air time tariffs (Rs 6 per minute).

Interconnections
Interconnection regulation had been in terms of interconnection charges (set up charges) and revenue share (usage charges)a . The basic framework was laid down in TRAIs consultation paper on Telecom Pricing (9 September 1998) and TTO 99. Key aspects of the framework were: Interconnection prices were based on costs and usage charges were based on a percentage of revenue share. For interconnections between fixed services providers, the provision for the local calls was on the basis of bill and keep, for NLD calls the revenue share proportion was 40:60 for the originating and terminating service provider, respectively. For ILD services, revenue share proportion was 45:55 between the originating and terminating network, in this case, the DoT. For interconnection between fixed and mobile services, the Receiving Party Pays principle was followed, with no revenue share. The called party would pay the airtime. For calls between the mobile to fixed services, the originator would pay the airtime as well as the Public Switched Telephone Network (PSTN) charges (as applicable). The PSTN charges were to be passed on to the fixed service provider in toto for the long distance and the international component. TRAI proposed a shift to the CPP regime in September 1998 but shifted the implementation to August 1999, as the implications of the National Telecom Policy 1999 (NTP 99) that was likely to be announced in March 1999 also needed to be factored in. The NTP 99 changed the annual licence fee to a 1-time entry fee and an annual revenue share. This had implications on the cost-based tariffs. TRAI subsequently reviewed the tariffs and reduced the cap on rentals (from Rs 600 to Rs 450) and airtime (from Rs 6.00 per minute to Rs 4.0 in metros and Rs 4.50 in circles). The revenue share for NLD and ILD continued as before. For fixed to mobile local calls, TRAI specified a charge of Rs 2.40 for the first minute and Rs 1.20 for each successive minute. Along with it, the revenue share was mandated as 33:67 per cent between the fixed and mobile operator specified as a mobile terminating charge. Due to MTNL and others filing a case in the High Court requesting for a stay on the CPP regime on the grounds that this would lead to (i) increased costs due to network upgradation, bill collection, and bad debts, and (ii)TRAI having no jurisdiction to issue or to make regulations to regulate arrangements amongst service providers. The courts held that TRAI did not have powers to alter the terms and conditions of the licence (through specifying the revenue sharing regulation). Subsequently, the government issued the TRAI Amendment Ordinance 2000 in January 2000. This changed the composition and powers of TRAI, specifically giving TRAI the power to fix interconnectivity terms, and setting up the Telecom Dispute Settlement Appellate Tribunal (TDSAT). In addition to the scope of the disputes in the earlier act, the tribunal would also be the appeal mechanism

A Review of Telecom Regulatory Authority of Indias Tariff and Interconnection Regulation 241
for the decisions of TRAI. The decisions of the tribunal could be appealed against only in the Supreme Court. (for a detailed analysis, see Sinha 2001)).

THE SECOND PHASE


The second phase of the rebalancing was to be effective from 1 April 2000 but was postponed to July 2000, as due to the lag in the availability of data for the required period of operation, the TRAI could not do the necessary review. This was further postponed by a month to August 2000. During the review process the service providing arm of the erstwhile DoT, called Department of Telecom Services (DTS), (now BSNL) mentioned that due to tariff rebalancing, (the long distance calls becoming cheaper), it had earned lower revenues by Rs 20002200 crore during 19992000. The flexibility of the alternative tariff package had been utilized to the extent of not charging rent for subscribers making 200 MCUs. This had led to a further loss of Rs 10001200 crore. However, in TRAIs assessment, due to the delays in implementation of the second round of rebalancing, declining cost of telecom equipment and overall revenue increase due to lower priced STS/ISD calls, the effects of the rebalancing in the first phase had not been very severe. As it expected the DTS to bring about efficiency changes to further reduce its costs, TRAI decided to go ahead with the second round of rebalancing which would be applicable until 31 March 2002. The longer time frame would enable elasticity of demand to manifest itself, and give greater time to the incumbent to make structural adjustments. The STD rates were expected to decline by nearly 11.5 per cent, while the rentals were not expected to significantly change during this period. While the original plan had envisaged an increase in rentals for the general subscriber category, the TRAI decided not to implement it, as it felt that the total loss on this account would be about Rs 200220 crore. This amount, as a percentage of total revenues of DTS was very small and could be made up by greater number of STD/ISD calls and improved efficiency of the incumbent. It felt that not increasing the rental would help in the increase of teledensity. Since TRAI was in the process of working out the cost-based tariffs on the basis of forward looking costs and a different methodology, both of which were going to be a part of a further consultation process, it deferred an increase in rentals. The revised tariffs were to be effective from 1 October 2000. In August 2000, the National Long Distance Competition policy that would allow private operators to offer long-distance services was announced and it was expected that this would bring down long-distance prices. A revision in the cellular air time to Rs 3 per minute was also affected.
Set up Costs of Interconnection is specified as the initial cost of any engineering work needed to provide the specific interconnection facilities requested and the Usage Charge is specified as the charge levied by a service provider for carriage of telecommunication traffic on its network.
a The

would have to follow the associated roll out obligations. The cellular operators had regarded this as a backdoor entry for fixed service operators to provide cellular services, without having to pay the high licence fees for spectrum that they had paid in the auctions. The Cellular Operators Association of India (COAI) had taken this issue to the Telecom Dispute Settlement Appellate Tribunal (TDSAT), that had agreed with TRAIs decision to allow WLL (LM) services. The COAI had then appealed against the TDSAT decision in the Supreme Court. The Supreme Court had allowed the roll out of WLL (LM) services while enjoining the TDSAT to examine the issues of level playing field raised by the COAI. Any disadvantage to the cellular operators due to the regulation was to be neutralized. The COAI considered WLL (LM) services as unfair impositions in the market. Therefore, when Tata Teleservices, a fixed service provider, rolled out its CDMA-based WLL (LM) network in mid-January 2003, the cellular operators denied it interconnectivity with their networks. Since there was no formal interconnect agreement of the fixed service operators with cellular operators, Tata Teleservices started routing its calls through the BSNL/MTNL network. But the cellular operators were able to block such calls too. In

retaliation, BSNL and MTNL cut off access to selected cellular operators. Subsequently, the then Minister for Information Technology and Telecommunications, Pramod Mahajan, after a meeting with cellular operators, announced that he would be willing to arbitrate in the dispute to ensure fair and equitable outcomes and all networks should conform to the need for interconnectivity with others.

THE THIRD PHASE REGULATION

OF

INTERCONNECTION

Since the March 1999 consultation dealt only with interconnections of the various service providers with DoT, TRAI brought out a consultation paper on 14 December 2001 to address the interconnection issues and to develop a General Framework for Interconnection (GFI) in the context of private NLD operators entry into the telecom service market. This provided a methodology for charging carriage of a long distance call in a multi-operator environment and to discuss issues relating to equal ease of access by subscribers to the NLD networks particularly relating to Carrier Access Code (CAC), pre-selection, and default carrier (http://www.trai.gov.in/intpap.doc).

242

India Infrastructure Report 2004 services covering basic, including WLL (M), services, cellular mobile service and long-distance services. The interconnection charges would continue to be governed by the Telecommunications Interconnections (Charges and Revenue Sharing) Regulation 2001, except to the extent that it was modified by the current regulation.

Subsequently, consultations with the stakeholders as well as the TRAI consultation paper 2001/5 dated 14 December 2001 led to regulation for interconnection charges and revenue sharing for WLL (LM). These were as per the fixed service licences. TRAI felt that given the changed market conditions (introduction of greater competition in cellular, introduction of players offering WLL (LM), and competition in the NLD and ILD segments) in the interim, the tariffs for other categories needed to be reviewed. By this time the long-distance tariffs had fallen considerably (and in excess of those stipulated in the TTO 99), due to the introduction of competition. Annexe 10.2 gives the existing NLD calls as well as those specified in the TTO 99.

Access Deficit
Since the first tariff review, while the NLD and ILD tariffs had been substantially reduced, adequate increments in rentals and local charges had not taken place. As per TRAI, this had created an access deficit for the fixed service provider. This was due to rentals and local call charges continuing to be below cost. The tariffs in the TTO 2003 were, as earlier, based on an affordability criterion and the objective of increasing teledensity. TRAIs studies (http://trai.gov.in/24thamend ment.htm) indicated that users would not like rentals and local call charges to be increased. On this consideration, TRAI did not increase rentals and local call charges. The access deficit could only be compensated through NLD and ILD charges. The ADC was to be recovered through IUC. Consideration of acceptable rentals based on Consumer Price Index increments was rejected because the increase could have implications with respect to the differential of rentals between fixed and WLL (M) and cellular services. As per TRAI, the rentals for fixed services were to be determined so that changes are not brought about in a manner which reduces the spread of the basic fixed line services called Plain Old Telephone Systems (POTS), which is considered an essential service in a developing country like ours. (www.trai.gov.in/consultbasicpaper.htm, page 36) The commercial users were to pay higher rentals, while tariffs for the non-commercial users was not changed in the interest of affordability and increasing teledensity. A new rental category was defined for senior citizens (definition was the same as for payment of income tax). The pulse duration for local charge was reduced to 2 minutes from the earlier three minutes. Given the competition in the NLD, there was forbearance, subject to a ceiling of Rs 8.40 per minute for peak time. On the ILD segment, there was total forbearance. The IUC included the ADC payable to the fixed service operators which they must get in order to keep the rentals as well as local calls affordable (http://www.trai.gov.in/ consul25.htm, page 6, point 1). The IUC determination was based on the detailed data given by the service providers based on an assessment of the various cost items attributable to the network elements involved in the different stages in setting up of a call in a multi-operator environment.

Events in 2002 and 2003


The third phase of rebalancing consisted of 2 parts. These were implemented through TTO dated 14 March 2002 and TTO dated 24 January 2003, respectively. In the TTO dated 24 March 2002, TRAI decided to club the categories of general and low user into a single category of noncommercial user subscriber. Since a tariff review was under way, TRAI decided to implement the tariffs as per the earlier order and only for the commercial user subscriber. The service providers were required to identify the commercial user subscriber and implement the tariff package accordingly. The free calls were reduced to 30 and 45 for the urban and rural commercial subscribers (Annexe 10.3). In July 2002, TRAI came out with a framework for a Reference Interconnect Offer (RIO). The RIO specified the framework and the unbundled network elements to be taken into account while calculating the IUC, that was applicable in a multi-operator context. The players with significant market power (players with more than 40 per cent of the relevant market share) were required to publish the RIO. TRAIs consultation paper on Tariffs for Cellular Mobile Telephone Services dated 8 July 2002 (www.trai.gov.in/ cellular%20cons%20paper-final.htm) related to the changes in cellular tariffs that had happened because of changes in the cost structure of the cellular operator due to decreasing costs and increasing subscriber base. A review of the competitive trends in basic services was undertaken by TRAI (http://www.trai.gov.in/consultbasicpaper.htm) in September 2002. This paper laid out the basis of tariff rebalancing and the possibility of using ADC. It laid out the framework for estimating cost-based tariff, various means of addressing the access deficit, and mechanisms to estimate the IUC. TRAI came up with the TTO 2003 and the accompanying Telecommunications Interconnections Usage Charges Regulation. These formed the basis for the third phase of rebalancing. The regulation covered arrangements amongst service providers for payment of IUC for telecommunication

A Review of Telecom Regulatory Authority of Indias Tariff and Interconnection Regulation 243

ADC Computation
For Cellular Mobile and Wireless in Local Loop with limited mobility (WLL-M), the Access Deficit Charge is not applicable as the rentals and call charges have been left to market forces and have not been kept below cost by regulation, as is the case with fixed local telephone service (POTS) (www.trai.gov.in/consul25.htm, page 8, point 8). For fixed services, the ADC was assessed by fixing an affordable level for rental/ local call charges, such as monthly rentals, local call charges, special concessionary local call charges in the rural areas, provision of free calls, and other below cost tariffs that the regulator specified to make the fixed services affordable to the common man to promote both Universal Service and Universal Access as per NTP 99 rather than leave these to the market. In order to reach the final estimates of IUC this Regulation takes into account the requirements of Access Deficit Charge arising out of the Tariff Order being issued (www.trai.gov.in/consul25.htm, page 8, point 7). For calculating the IUC, cost data from BSNL and MTNL, which are the dominant operators providing fixed and NLD services, was taken. For deriving the capex, TRAI took the audited figures of BSNL, for the financial year 20012 (Annexe 10.4). The components included data on depreciation charges during the year, net block, capital works in progress, current assets, current liability, employees remuneration and administrative expenses. Information on the number of DELs at the end of the period was also provided. CAPEX + Depreciation costs and OPEX costs had been converted to cost per line against these heads by dividing the costs with the DELs as on 31 March 2002. The derived values were adjusted for costs attributable to telephone service with the assumption that only 95 per cent of total revenues are derived from these services (Annexe 10.5). The overall CAPEX and OPEX were allocated to different portions of the network by allocating these costs in the same ratio as was done by the BSNL in its RIO Schedule report cost data submitted to the TRAI. Thus, based on BSNL inputs, the aggregate amount of CAPEX + OPEX had been allocated to network elements based on Mean Capital Employed for each un-bundled network element as shown in Annexe 10.6. On this basis, the rentals were calculated as Rs 424 per month. The average recovery of BSNL on the rentals was Rs 165175 per month. TRAI suggested that the access deficit on rentals amounting to Rs 249259 needed to be recovered through ADC. In order to calculate the total amount to be recovered through the ADC, the free calls and the 300 below cost metered call units (MCUs) provided in rural areas were factored in (Annexe 10.7). The ADC-incorporated local charges were estimated as Re 1 per minute, while those for long distance were Re 1

per minute plus the cost of long distance carriage. But since the per minute charges had been fixed at Rs 0.80, Rs 1.00 or Rs 1.20 per two minute duration, the portion of the access deficit not recovered from the local calls was to be recovered from the long distance charge. For the rental portion, the capex up to the SDCC, licence fee (revenue share), and spectrum charge (12 per cent) were taken. Local charges were taken on the opex for that part of the segment distributed across the minutes of usage. A similar principle of relevant costs (based on the work done) distributed over minutes of usage for computing the long distance charge was used. In this case licence fee (revenue share) was taken as 15 per cent. Both origination and termination charges for fixed line to fixed-line calls were taken as equal. For the long distance charge, the access deficit on account of below cost local charge was also added. It was also decided that the IUC (including the ADC component) would be uniform (independent of the distance slabs) due to current technologys inability to implement differential IUCs. With the ADC loaded to the IUC, the new NLD rates were higher than the then prevailing NLD prices. The fixed to cellular calls would be chargeable at Rs 1.20 for 90 seconds in metros and Rs 1.20 for 60 seconds in circles. For cellular services, a mobile termination charge, based on costs of termination was specified. With this, the receiving party was not required to pay for the incoming calls. In effect, this implemented the CPP. The cellular operator would get Rs 0.30 as mobile termination charge (MTC). The balance would be retained by the fixed service provider. For intra circle calls, the charge was Rs 1.20 per minute and the cellular operator would get Rs 0.40 per minute for termination. For cellular to fixed line, the cellular operator would give Rs 0.50 in metros and Rs 0.60 in circles, respectively. An additional Rs 0.20 would be paid to transit Level II tax in circles. These charges were based on the existing subscriber base, a quarterly annual growth of 25 per cent for January to March 2003 and an annual growth rate of 70 per cent over the period April 2003 to March 2004. The opex was based on data of 25 metro and circle operators based on the annual audited figures reckoned to work out opex as on 31 March 2004. The per minute usage figures of 220 minutes was expected to go up to 250 minutes (Annexe 10.8). Unlike the earlier CPP introduction, this time TRAI felt that the additional costs to be borne by the fixed subscriber (to pay for the MTC) were not high. The MTC had come down due to fall in prices of the network elements, increasing subscriber base and additional revenue from value added services (VAS). On the plus slide, the TRAI felt, it would enhance the call completion rates as the called party would not have an incentive to keep the handset off to avoid unwanted calls.

244

India Infrastructure Report 2004 them over fixed line. Creating price differentials through regulation could create demand for a particular technology that would not otherwise exist in the market. Such interventions would be detrimental to the growth of the sector. Technology choices should be left to the market.

The changes in tariffs were viewed as significant due to the shorter call duration, reduction in number of free phone calls (from 75 to 30 for urban), and what was viewed as differential pricing for BSNL/MTNL subscribers making calls to cellular/WLL(LM) phones. While earlier, a 3-minute call from a landline in a metro to cellular subscriber in the same service area cost Rs 1.20, the same would cost Rs 3.60 and a call to WLL(LM) would cost Rs 2.40.

The Extent of Tariff Rebalancing


While TRAI started out with a mandate for bringing about full tariff rebalancing, its performance in this area has been poor. The consultation paper on Tariffs for Basic Services (23 September 2002, page 24) notes in the context of rebalancing, the objective would be to make the access deficit zero by raising the rental/local call charges to their cost-based levels. In the following sections we identify the major issues that needed to be resolved for achieving the objective.

The Political Fall Out


Due to political pressure and the media coverage of this issue, the then Minister for Information Technology and Telecommunications, Arun Shourie, was summoned for a meeting with the Prime Minister, the Deputy Prime Minister, the Union finance minister, and the BJP president, M Venkiah Naidu to discuss the political fallout of the tariff increase. This led to an announcement of a partial roll back of the tariff increase. The number of free calls for BSNL subscribers was raised to 50 and 75 for the urban and rural categories, respectively. The MTNL subscribers would get 60 free calls instead of the 30 announced earlier. The fixed to cellular calls anywhere in the country would cost Rs 3.60 for 2 minutes. TRAI also announced it would review its regulation on the implementation of the IUC and sought comments and held open house discussions as there were several concerns regarding the treatment of ADC, its computation, and its incorporation into the IUC.

The Segmentation Framework


The TTO 99 gave the right perspective by segmenting subscribers as urban and rural and, within these categories, as low, general, and commercial, and attempted to fix the rentals accordingly. However, the basis for categorization that was used for low users (MCUs) was not carried through for general and commercial categories. In addition, the rentals for the different categories did not indicate a clear distinction between the subscriber categories (Annexe 10.1). For example, a) the rentals for the low and general user in the rural and urban categories were kept the same in the first phase (from 1 April 1999 to 31 March 2000), b) the rentals for the general and commercial categories both in the rural and urban segments were kept the same during the third phase (1 April 2001 to 31 March 2002), c) TRAI clubbed the low and general categories into one category non commercial, and d) after the first increase in rentals, TRAI did not plan to increase them for the low user and commercial categories during the subsequent periods. Thus, TRAI was unable to leverage the differences in the requirements for the 3 categories.

THE ISSUES

AND AN

EVALUATION

We analyse the organizational and policy issues in the context of this exercise of tariff rebalancing and subsequently focus on issues relating to the treatment of ADC and IUC.

Independence of TRAI
Despite the TRAI Amendment Ordinance (2000), that gave it the power to fix interconnectivity terms, TRAI has not been able to enforce its mandate as is evident from the case of cellular operators denying interconnections to WLL(LM) operators or BSNLs blocking of calls. By not enforcing the interconnections, TRAI created space for intervention by the minister.

Basis for Increasing Rentals


TRAI could have been more aggressive about increases in rentals, as in the years during the tariff rebalancing, the Consumer Price Index had been increasing at 10 per cent per annum. But it based its decision to a general sense of affordability. While, despite the increase in consumer prices, it may have been difficult to raise rentals for the low user, the potential of commercial users to pay higher rentals was not fully exploited. This was possibly not done as commercial consumers may have demanded higher rentals to be linked to better quality services (for example, flexibility in billing, payments, etc.) which DoT/BSNL may not have been able to commit to.

Technology Neutrality
While TRAI professes to be technologically neutral, its perspective that POTS through wireline is best for a country like India does not reflect neutrality. In another instance, exemplifying the same perspective (and contradictory to its stand on ADC), it did not recommend increasing wireline rentals as they would then compete with cellular and WLL(LM) rentals, and, therefore, subscribers could prefer

A Review of Telecom Regulatory Authority of Indias Tariff and Interconnection Regulation 245

Exploiting Alternative Tariff Packages


Even though TRAI was not very aggressive about the tariff rebalancing, DoT/BSNL, should have easily foreseen the emergent competition in NLD and offered alternative tariff packages to the subscribers much earlier.The consultation paper on Tariffs for Basic Services (23 September 2002, page 24) notes that: It is pertinent to mention that while rebalancing did allow for recalibration of commercial users rentals, none of the service providers have raised these rentals. The service providers have thus not rebalanced this element although they had the opportunity to do so, and have foregone some much needed resources..... We estimate the extent of loss in revenues due to not increasing the rentals. In March 1999, DOT/BSNL had nearly 16 million DELs in non-metros (about 70 per cent of them are urban, as per DoT/BSNL annual reports), and 5 million lines in metro cities which went up to a total of 31 million lines and 6.8 million, respectively, by 2002. A one-time conservative increase in the rentals of Rs 100 per month in the commercial category would have yielded additional revenues of Rs 1200 crore in 1999 going up to Rs 2000 crore in 2002. (This assumes that in the urban areas, 60 per cent lines are in commercial categories). This calculation does not take into account the possibility of higher revenues that could be generated by linking higher rentals to lower per minute call rates, leading to greater usage. While operators had the option of alternative tariff packages since 1999, they had not exercised this option until May 2003. If the fixed service providers do not leverage the flexibility (alternative tariff packages), they cannot claim that they were not allowed to recover access deficit.

Before identifying the specific set of subscribers who should receive subsidized access, there could be a process of self-selection. The targeted subscribers for whom affordability is an issue, could have a cap on usage charges. The total bill for such customers can be generated with lower rentals and lower call charges. The problem in such an approach is that it could lead to a situation where subscribers opt for multiple phones at lower rentals. This could be addressed by requiring a proof of residence and having a searchable database of addresses. In any case, this is the basic information that is required for billing purposes and is not a huge additional cost.

The Incorrectness of Data Values Supplied by BSNL


The values used by TRAI from the balance sheet of BSNL should only be used for creating some possible scenarios and not as a basis for decision-making. The balance sheet has several underlying significant assumptions that are not explicitly considered in the paper. For example, the basis of arriving at the BSNL valuation at the time of corporatization and, hence, the value of Gross Block and depreciation was historic costs. Since BSNL did not have commercial accounting practices, there were problems with this valuation. Various other examples of problem areas are provided in Annexe 10.9. The data problems are likely to remain unless BSNL adopts commercial accounting practices. Other examples of discrepancies are: The Gross Block numbers used to calculate the capex also reflects the investments made by BSNL for its cellular services. As per the annual report a capacity of 4 million cellular lines have been installed, of which 2.4 million lines have been made operational. These amounts need to be deducted for calculating the fixed cost per line. While BSNLs ROR on capital employed is 9.6 per cent, in calculating the pre-tax percentage for allocation to capital, it has been taken as 13.78 per centfor a regulated industry these are far too high2. The values of net block have been arrived using the historical cost. These reflect the inefficiencies of the past, and do not take into account falling prices, at least in some segment (exchange) centres, for land lines.

Freedom in Bundling Targeted Subsidies


Lower rentals need to be targeted only for those customers who cannot afford to pay the true cost of service provision. TRAI based its calculations of lower rentals on a study which fixed them according to the level of affordability for different categories. However, the TRAI consultation paper does not provide a reference to the study and the underlying assumptions. In contrast, a number of various other studies have shown that the rural people spend significant amounts on telecom services. The trade-off for them is in terms of cost of alternative means of communications (Jain and Sastry, 1997, 1999). Categorization of all subscribers who live in rural areas as qualifying for lower rentals and higher free calls is against any economic principles. This leads to poor services in rural areas (as operators are unable to recover costs), further exacerbating the problem. The rentals should be based on MCUs and subsidies should be targeted.

Treatment of Access Deficit


Much, therefore, depends on the flexibilities available in the existing set of tariffs, that is, those relating to the NLD and ILD sectors for rebalancing. The current paper would need to factor in the changed competitive conditions
Indeed, for service providers, which are trying to take advantage of network economies, by rapid growth of its network, strategic pricing would, at that stage, mean low return.
2

246

India Infrastructure Report 2004 The TRAI consultation paper assesses the ADC by adjusting the fixed cost per line per year as Rs 5120.46 (Annexe 3.5), which has a pre-tax return of 13.48 per cent built into it only against the rental of Rs 200 per month plus the cost of free calls. However, it should also take into account the revenue per line (Rs 7317 per year) after adjusting for the opex of Rs 2243.34 per DEL per year. (As per BSNL Annual Report: Total income from services Rs 242,998,944 thousand, total DELs 332184198, which gives an income per DEL per year as Rs 7317.) Since the actual pre-tax return on capital for BSNL is 9.36 per cent, the actual revenue generated by BSNL more than adequately covers the capex and opex.

as well as the feasibility and desirability of using IUC as a means to address the issue of access deficit. After stating this, the consultation paper does not discuss the alternatives for managing the access deficit. It assumes that IUC is the best mechanism to manage it. There must be deliberations on whether it is indeed so.

USO as a Better Mechanism than IUC to Fund the AD


The IUC mechanism to implement the ADC is not appropriate. The loading of ADC implies that subscribers of all networks that interconnect to the BSNL/MTNL network, in effect, subsidize all the subscribers including the commercial and urban subscribers of the BSNL/MTNL network. While the USO mechanism imposes a cap on subsidy that is based on the revenues, the IUC (loaded with ADC) is determined on the basis of subsidy that has already been decided to be provided. The IUC (loaded with ADC) does not provide any signals regarding the quantum at which the ADC will be capped.

The Benefits to the Incumbent


Any access deficit calculation of revenue shortfall arising from regulatory constraints should be significantly discounted by a value ascribed to the strategic benefits that BSNL derives from basic access. Being the incumbent and dominant provider of all telecommunication services, it has certain advantages that may balance the costs associated with access. These are: a) as the provider of basic access, BSNL is necessarily the preferred supplier of all call services. For example, users must make a deliberate choice for a competitive service provider, which may involve operational shortcomings (for example, use of a conditional access code); b) BSNL is able to include all services on one bill; c) being the incumbent and the universal service provider brings BSNL brand recognition; and d) as an incumbent BSNL has certain cost benefits of network operations (for example, early acquisition of sites for exchanges, towers), established land access (for trenches), etc.

Inappropriate Unbundling of Costs and Revenues for Rentals and Free Calls
In calculating the ADC, TRAI unbundled the costs of the local network and free calls and reviewed them against the revenue attributable to rentals as opposed to calculating the total revenue in relation to the total cost. Economically, there is no reason to presume that an input that is used to provide a variety of different services must have its cost recovered from a particular charge. Rather, if one input is used to facilitate a large variety of retail products, then any evaluation of the deficit or surplus associated with that input needs to consider the entire operations of the relevant firm. From the subscribers perspective, in acquiring a connection is the trade-off in the various costs (rentals, local, long distance, international, Internet, cellular) balanced against the benefits. From the operators perspective, it is the collective revenue from all services that indicate the profitability of the specific customer. If providing access to a customer is profitable when all revenues and costs associated with that customer are considered, then there is no meaningful access deficit for that customer. But the access deficit has been calculated only with regard to rentals and free calls ignoring any economic profits that accrue to the fixed service providers including BSNL from other telecommunications services. The availability of other services, such as cellular and Internet over the same network gives rise to additional calls, the revenue for which accrues to the access provider as these calls originate at the subscriber end. For example, the cellular to fixed line and fixed to cellular calls constitute 70 per cent of the total cellular calls.

Desegregated Data for Different Service Components And Accounting Separation


The revenue/DEL used by TRAI includes the long distance and other revenues. The only way to arrive at desegregated costs and revenues of the various segments is probably through accounting separation of BSNL/MTNL into BSNL Cellular, BSNL Long Distance, and BSNL Access. These units would have to then offer their products and services at non-discriminatory prices to any external party vis--vis the various BSNL units. In such a scenario, if indeed BSNL access has a deficit, then there could be a case for IUC (with ADC).

Performance-Based ADC
While BSNL has justified its high long-distance charges due to the social obligations of having to meet rural demand, the recent (C&AG) report has identified that nearly 45 per

A Review of Telecom Regulatory Authority of Indias Tariff and Interconnection Regulation 247 cent of the village phones do not work. Similar results have been reported in other studies (Jain and Sastry 1997, 1999). Thus, BSNL has not only foregone revenues from call origination but also from those phones in other parts of the network wishing to contact the people in villages if the phone had been working. During the monopolistic regime of DoT/BSNL, the spread of the rural network had been poor. Even today, only 25 per cent of its DELs are in the rural areas. This implies that without explicit performance criteria, it would be difficult to ensure adequate rural coverage. The ADC recovery through IUC does not provide for such a mechanism. The quantum of deficit is first arrived at and then built into the IUC. Even if the ADC-based IUC is to be implemented, the ADC calculations should vary over the implementation period, and provide for giving incentives to the compensated firms to bring down the capex and opex values. For example, TRAI should specify the percentage points by which the capex and opex values used in estimating ADC would come down over a specified period. The TTO 2003 and the accompanying regulation should also specify the time when the next review would be undertaken. The USO consultation paper has provided a framework for assessing the access deficit (Annexe 10.10). Thus, the only significant issue to be examined is whether the quantum provided in the USO consultation paper is adequate, and if not, best to increase it. It could be through an increase in the percentage revenue of licence fee towards the USO (say from 5 per cent to 7 per cent). Since the increment is envisaged to come out of the existing licence fee, no additional administrative changes need to be worked out to manage a larger corpus. business will constitute only 60 per cent of its business by 2007. The growing size of cellular networks (both the subscriber base and minutes of usage) would ensure that revenues contributed by the cellular subscriber would be very significant. An increasing number would come from the non-urban areas. However, the ADC is based on fixed networks. If using fixed networks becomes more expensive, (because of the addition of ADC component of IUC), this would cause a further migration to cellular networks (especially as, by then, cellular networks would have reached significant proportions of the total network size).

CONCLUSIONS
This chapter raises issues about the role of TRAI in the context of facilitating competition through tariff rebalancing and interconnections. While TRAI began the tariff rebalancing exercise with the objective of completing the same within 3 years, the target was ambitious to begin with. Subsequently, it was not able to maintain the pace. A more realistic timeframe would have led to a greater certainty and credibility. TRAI lost significance of its role when it was not able to enforce interconnections. Its perspective on assessment of the ADC reflects the DoT/BSNL viewpoint rather than an unbiased approach. The manner in which TRAIs consultation papers on tariff regulation have segmented the customer reflect a lack of technical expertise. TRAI would need to ensure that its decisions reflect technology neutrality. Otherwise the sector could see distorted growth due to regulatory interventions. For this to happen, a change in the mind-set of those responsible for making such decisions is critical.

Incorporating the Future Trends: Fixed Line and Wireless Networks


The growth in wireless is expected to be extremely fast, especially in the next 510 years. For example, during 20027, only 18 per cent of BSNLs future additions are going to be in the fixed line category. The remaining 82 per cent would be in the cellular and WLL categories3. The fixed line that currently constitutes 95 per cent of BSNL
As per a news item in Economic Times, New Delhi: The public sector telecom companies Bharat Sanchar Nigam Ltd and Mahanagar Telephone Nigam Ltd would add about 3.95 crore telephones by 2007. In the current 5-year plan (20027), BSNL would add 81.70 lakh landline phones, 2.18 crore mobile phones and 68 lakh Wireless in Local Loop (WLL) phones, meaning thereby that the number of mobile phones added by BSNL would far exceed fixed phones and WLL phone additions. The financial outlay for BSNL has been earmarked at Rs 66,412 crore, an official release said here. Another state-owned corporation MTNL would add 16 lakh landline and WLL telephones and 11.57 lakh
3

mobile phones during the plan period. The private sector is expected to add 2.55 crore Direct Exchange Lines (DELs) during the period, it added.

248

India Infrastructure Report 2004


Annexe 10.1 Monthly Rentals for Different Subscriber Categories (TTO 1999) (Rs)

Urban Subscribers Exchange System Capacity Number of Lines Low 1000 to 29,999 30,000 to 99,999 1 lakh and above Rural Subscribers Up to 999 1000 to 29,999 30,000 to 99,999 1 lakh and above 70 120 180 250 70 120 180 250 120 160 220 310 70 120 180 250 95 140 200 280 120 160 220 310 70 120 180 250 120 160 220 310 120 160 220 310 120 180 250 From 1 April 1999 to 31 March 2000 Type of User General 120 180 250 Commercial 160 220 310 Low 120 180 250 From 1 April 2000 to 31 March 2001 Type of User General 140 200 280 Commercial 160 220 310 Low 120 180 250 From 1 April 2001 to 31 March 2002 Type of User General 160 220 310 Commercial 160 220 310

Source: TRAI, The Telecommunication Tariff Order 1999, pages 912.

Annexe 10.2 STD Call Charge for Fixed to Fixed Calls (Call duration of 1 minute and pulse charge Rs 1.20 per metered call) Distance category Peak tariff envisaged at end of tariff rebalancing under TTO 1999 (1 April 2002) 1.2 4.8 10.8 16.8 21.6 Prevailing rate at present Percentage reduction

Intra Circle Up to 50 km 51200 km 201500 km 5011000 km > 1000 km 1.2 2.4 2.4 2.4 2.4

Inter Circle 1.2 2.4 4.8 4.8 4.8

Intra Circle Nil 50 78 86 89

Inter Circle Nil 50 56 72 78

Source: TRAI, Consultation Paper on IUC issues, 15 May 2003, page 39.

Annexe 10.3 Monthly Rentals for Rural and Urban Subscribers (Rs) Exchange System Capacity Number of Lines Up to 999 1000 to 29,999 30,000 to 99,999 1 lakh and above 70 120 180 250 Rural Type of User Non-Commercial Commercial 120 160 220 310 120 180 250 Urban Type of User Non-Commercial Commercial 160 220 310

Source: TRAI, The Telecommunication Tariff Order, 14 March 2002.

A Review of Telecom Regulatory Authority of Indias Tariff and Interconnection Regulation 249
Annexe 10.4 Profit and Loss Account for the year ended 31st March 2002 (Rs 000s) Year ended 31 March 02 Income from Services Other income Total Expenditure Employees Remuneration and Benefits Licence fee and spectrum fee (Ref. note 13, 14 on schedule U) Administrative, Operating and other expenses Financial expenses Depreciation Total Profit before prior period adjustment and taxation Prior period adjustments Profit before taxation Provision for taxation Profit after taxation Appropriation: Bonds Redemption Reserve Surplus carried to balance sheet Earnings per share Basic/Diluted earnings per share (Rs) Refer note 20 on Schedule U Source: Annual Report 20012, BSNL. 242,998,944 26,817,995 269,816,939 38,484,520 34,031,191 39,957,915 4,682,106 87,461,309 204,617,041 65,199,898 3,321,938 68,521,836 5,400,141 63,121,695 5,719,018 57,402,677 63,121,695 12.62 Year ended 31 March 01 115,966,611 1,028,156 116,994,767 20,700,739 15,732,466 28,937,314 2,742,899 38,580,811 106,694,229 10,300,538 10,300,538 2,830,000 7,470,538 3,717,746 3,752,792 7,470,538 1.49

Annexe 10.5 Calculations of CAPEX, Depreciation and OPEX Components per DEL based BSNL Cost Data for Year 20012 Total Number of DELs as on 31.3.2002 (in crore) As per BSNL audited figures on 31.3.2002 (Rs in crore) Depreciation charged during the year Depreciation per line in Rs Net Block Capital Works in progress Current assets Current liability Amount to be considered for multiplication by pre-tax weighted allocation of capital Pre-tax weighted allocation of capital percentage CAPEX Component CAPEX Component per line in Rs CAPEX + Depreciation per line in Rs CAPEX + Depreciation per line in Rs attributable to telephone services (95 per cent) OPEX Employees Remuneration Administration Total OPEX OPEX per line in Rs OPEX per line in Rs attributable to telephone services (95 per cent) Source: TRAI, The Telecommunication IUC, 24 January 2003, page 30. 3.32 8746.13 2632.91 58,922.21 10,826.24 17,083.4 20,369.55 66,462.3 13.78 9158.5 2757.05 5389.96 5120.46 3848.45 3995.79 7844.24 2361 2243.34

250

India Infrastructure Report 2004

Annexe 10.6 Apportionment of CAPEX + Depreciation and OPEX to Different Network Elements Based on Audited accounts of BSNL for the Year 20012 based on the Mean Capital Employed by BSNL as given in Schedule 5 of RIO by BSNL Network Element (NE) Share of Mean Capital Employed per line (in per cent) 54.78 20.38 1.27 1.25 1.25 3.26 7.74 0.66 3.85 0.31 1.64 0.33 3.27 100.00 Annual CAPEX + Depreciation as apportioned for Network Element excluding licence and spectrum fee 2805 1044 65 64 64 167 397 34 197 16 84 17 167 5120 Annual OPEX as apportioned for Network Element 1229 457 29 28 28 73 174 15 86 7 37 7 73 2243

Access Loop Local Exchange SDCC Tandem Intra-LDCA (Level II Tax) Inter-LDCC Intra-circle + Inter Circle (Level 1) LE-SDCC Transmission System LE-SDCC Transmission Length (Avg. 10 km) SDCC-LDCC Tax Trans. SDCC-LDCC Tax Transmission Length (Avg. 40 km) Inter-TAX Transmission Length (Intra-Circle) [SDH Rings] Inter-TAX Transmission Length (Inter-Circle) Inter-TAX Transmission SDH-16 System (Inter-Circle) Inter-TAX Transmission Length (Inter-circle) [SDH Rings] Total

Source: TRAI, The Telecommunication IUC, 24 January 2003, page 31.

Annexe 10.7 Access Deficit Estimation No. of fixed subscribers 40 Average cost-based rental Rs 425 Average rental actually charged Rs 200 Deficit per fixed phone per month Rs 225 Annual deficit (per fixed line) Rs 225 12 = Rs 2700 Annual deficit on account of rentals for 40 million fixed subscribers Rs 10,800 Average number of free calls 30 per subscribers per month Rs 1440 Deficit on this account Deficit on account of below cost calls between 0 to 50 km (706 calls per subscribers per year) Per call deficit 25 paise per call Rs 750 Total annual access deficit estimate Rs 13,000 Source: TRAI, Consultation Paper on IUC issues, 15 May 2003, page 40. million per month

crore crore

crore crore

Annexe 10.8 Illustrative IUC Charges and Prevailing Tariff for Different Type of Calls >500 km Inter Circle F-F F-W F-C W-F W-W W-C C-F C-W C-C 5.1 3.6 3.5 3.6 2.1 2.0 3.5 2.1 1.9 Intra Circle 5.1 3.6 1.2 3.5 2.0 1.0 1.2 1.0 0.8 4.75 3.25 3.15 3.25 1.75 1.65 3.15 1.65 1.55 200500 km Inter Circle Intra Circle 4.75 3.25 1.2 3.15 1.65 1.0 1.2 1.0 0.8 4.45 2.95 2.85 2.95 1.45 1.35 2.85 1.35 1.25 50200 km Inter Circle Intra Circle 2.45 1.95 1.2 1.85 1.35 1.0 1.2 1.0 0.8 0.5 0.85 0.75 0.85 1.2 1.1 0.75 1.1 1.0 050 km Inter Circle Intra Circle 0.7 0.95 1.2 0.85 1.1 1.0 1.2 1 0.8

Notes: F Fixed; W WLL(LM); C Cellular. Source: TRAI, Consultation Paper on IUC issues, 15 May 2003, pages 434.

A Review of Telecom Regulatory Authority of Indias Tariff and Interconnection Regulation 251
Annexe 10.9 Examples of Problems with the BSNL Annual Report Data As per the auditors report The accounting policies regarding depreciation of fixed assets are not uniformly applied across all the circles. The depreciation has been provided on fixed assets taken over from DoT as of 1 October 2000 on written down value method at rates as per the Companies Act, as if it were its original assets and has not calculated the rates for the remaining useful life of the asset (page 48, BSNL, Annual Report, 20012). As per the auditors report, there are several places where it is mentioned that appropriate commercial policies have not been followed in computing the depreciation. This has implications for the value used in calculating the cost per line based on depreciation for the year (TRAI Consultation Paper 24 January 2003, Table 1, page 30). The overheads (Establishment Expenses) are allocated as a percentage of capital expenditure at percentages prescribed by DoT and not on the basis of directly allocable costs (page 50, Annual Report, BSNL 20012). At several places the auditors have indicated that accounting practices recommended by the Institute of Chartered Accountants is not being followed and amounts under several heads of account are unascertainable. The fixed assets have not been revalued during the year. It is not clear who (DoT, BSNL) will bear the incremental pension liabilities due to the pay increases on 1 October 2000.

Annexe 10.10 Relevant Excerpts from the USO Consultation Paper In its consultation paper on USO, TRAI had suggested an imposition of a Universal Service levy that included both access, which means public access through public or community telephone and provision of individual household telephones. Since the levy would support both these activities, it has been called TRAI Universal Service Levy (USL).. The Authority has recommended that implementation of USO should be divided in two clearly identifiable streams: First, for provision of public telecom and information services and, second, for provision of household phones in net high cost rural/remote areas. The Authority has recommended that support from Universal Service Fund (USF) be provided for Net Cost (that is, Cost minus Revenue) of providing VPTs/PTICs and DELs in rural/remote SDCAs. Details on the relevant costs are in the main recommendations. These include, for instance, no capital recovery for VPTs or phones installed before April 2002. For them, only operating expenses should be taken into account for estimating net cost. However, both capital recovery and operating expenses should be taken into account for VPTs, PTICs and phones installed after April 2002. Also, in a multi-operator environment, the lowest Net Cost computed by the proxy model for the least cost operator in an SDCA, should be used as the basis to compute USF support available to all operators. This figure that is, 5 per cent of the revenue of all the telecom operators, appears to be adequate to support Universal Service programme in its first phase of VPTs/PTICs as well as DELs in rural/remote areas. The amount of USL that is, 5 per cent of revenues, should come out of the licence fee itself and should not be an additional levy. Hence, the licence fee realized may be bifurcated into two parts. The designated portion of the Universal Service Levy may go the Universal Service fund and the balance to be Consolidated Fund of the Government of India. In subsequent years, the Universal Service Administrator may revise this figure depending upon the requirement.

REFERENCES
Jain, Rekha and Trilochan Sastry (1999) Assessment of Socioeconomic Impact of Rural Telecom Services: Implications for Policy, Workshop on Telecom Policy Initiatives: The Road Ahead. Jain, Rekha and Trilochan Sastry (1997) Rural Telecommunication Services, Workshop on Telecom Policy Research. Sinha, Sidharth (2001) Regulation of Tariffs and Interconnection: Case Studies, India Infrastructure Report 2001, Oxford University Press, New Delhi, India. Jain, Rekha and Dheeraj Sanghi (2002) Untangling Wireless in Local Loops, in 3iNetwork (2002), India Infrastructure Report: Governance Issues for Commercial, Oxford University Press, New Delhi.

You might also like