TELECOM TARIFFS IN INDIA
1. This section begins with a summary of the salient features of telecom tariffs in India, and then discusses some tariff issues which could be taken up for more immediate consideration. This is followed by a description of some important characteristics of the Indian telecom situation which bear upon any consideration of a general change in the telecom tariff policy. Finally, options regarding the telecom tariff policy in India are addressed.
IV.1 Some Salient Features of Telecom Tariffs in India
2. Telecom tariffs in India include a deposit at the time of registration, a non-reimbursable fee for initial connection, bi-monthly rentals whose amount depends on the capacity of the local network size, charges for local calls (in excess of free calls), charges for national and international calls, charges for manual trunk calls, and miscellaneous charges (including one years rental being kept as deposit).
3. Two systems for charging tariffs are used in India, a flat rate and a measured rate. Under the former, the subscriber has to pay a fixed monthly rental irrespective of the number of calls made. In the latter, the subscriber is charged a lower rental plus a usage charge per call. A specified number of calls are allowed free. At present 250 free calls per billing period are provided for rural subscribers, and non-rural subscribers are provided 150 free calls. Thus, the average number of free calls per day range from two and a half for non-rural subscribers to about four for rural subscribers.
4. STD calls are metered and charged according to a specified formula based on the time when the call is made and the distance covered by the calls.
5. Indian telecom tariffs have normally focused on revenue and social objectives, while incorporating certain cost considerations in the pricing methodology. A higher tariff is charged for covering a longer distance, calls at peak times are charged more, as are calls of a longer duration. A noteworthy feature in this context is that with modern technology costs have become largely independent of distance. Instead, costs depend more on the volume of telecom traffic. This is reflected, for example, in the tariff structure in OECD countries (Chart 6).
6. Indias tariff structure shows a sharp rise in tariffs as distance increases, with the highest rate for national long-distance calls being 90 times the lowest rate (see Table 4 and Chart 7). Moreover, there are escalating tariffs, i.e. higher the number of calls made by the user, higher is the tariff (Table 5). There is also a significant difference between tariffs for peak time and non-peak time usage (Tables 6 and 7, and Charts 8 and 9). Table 10 shows the pulse rates applied to international calls (peak and off-peak times), and indicates that the international call rates for most countries are higher than the upper limit of the tariff for domestic long-distance calls.
7. Discounts are offered for certain categories of customers. For example, lower tariffs are applied for rural subscribers for number of calls ranging between 150 to 450 (Table 5 and Chart 10).
8. Rental is subsidized in general. Even the highest rental charged is significantly below the estimated rental that will cover cost. Informal estimates suggest that if the current average rental is increased to cover the relevant cost, it would have to rise by 100 to 200 per cent.
9. The rental rates in India differ on the basis of the capacity of the subscribers exchange (Table 8 and Chart 11). At present subscribers in India linked to exchanges with lower capacity are charged a lower rental, and in the case of exchange capacity from 100 to 999 lines the rental for rural subscribers is less than that for others (Table 8 and Chart 11).
IV.2 Some issues for more immediate consideration of present tariff structure
10. This section addresses four aspects of the present tariff structure: escalation in tariffs; the pattern of tariffs for national long-distance calls (STD and operator-assisted trunk calls); and off-peak time tariff rates.
(a) Escalating tariff linked to the number of calls made
11. Table 5 shows the current practice of applying escalating tariffs as the number of calls increase. Such pricing discourages subscribers from making more calls, and encourages them to add another telephone instrument if the aggregate number of calls approach a relatively high figure. Moreover, the escalating tariff structure with five different components is not very transparent. Further, a rationalization of these tariffs could result in lower billing costs. Thus, there is a need to consider whether the prevailing structure of escalating tariffs can be replaced by a more simplified one.
12. A decision to change the tariff structure would also involve a consideration of whether to continue providing free calls. Another aspect to consider would be whether volume discounts should be provided to encourage subscribers to increase the number of their calls: in this context however, a relevant consideration is that wherever a capacity constraint is binding, the discount is unlikely to affect the number of calls made (Annex 2).
13. When considering a change in the tariff structure, it would be useful to also consider whether there is a need to change the basis of timing the duration of a call for the purpose of charging for that call. For example, should an initial flat rate tariff be charged for a specified time, followed by a per second charge applied to usage time which exceeds the time period for which the flat rate applies.
(b) Tariff structure for calls covering different distances
14. The present tariff structure for national STD shows a sharp increase in tariffs as distance increases (Table 4 and Chart 7). As mentioned above, with the advent of new technology, costs do not increase much after a particular distance is covered; and this distance is substantially less than the range that is specified for the Indian tariff structure: see for instance, Chart 6 which provides the OECD tariff structure linked to distance, and compare it to Chart 7 which shows a similar information for India; this comparison also has implications for a reconsideration of the international call tariffs. Thus, there is a need to consider whether this tariff structure should be altered to correspond more closely to the difference in costs that arise with a change in distance. Similarly, a related question is whether the tariffs for international calls should be oriented towards costs.
15. Cost-orientation of national STD tariffs would require reducing the distance for which the maximum tariff rate applies, and determining the various tariff rates for different distances on the basis of costs. If the national STD tariff structure were to be changed to better reflect costs, it would be necessary to also consider whether to retain the discrete distance-slabs as in the present structure of tariffs, or to use a more continuous and smooth increase in tariffs as the distance increases. If the distance-slab system were to be retained, even then one would need to determine how many slabs (and over what ranges of distance) should be considered for any revised tariff system.
(c) Comparison of national STD and operated-assisted trunk call rates
16. Table 9 compares current national STD tariffs and the tariffs fixed for operated-assisted trunk calls. Though the distance slabs are largely similar, there is a dis-similarity in the charging pattern: Table 9 shows that there is no consistent structure regarding the difference between the national STD tariffs and the operator-assisted trunk call tariffs over different distances. Hence, we need to consider whether the tariffs for operator-assisted trunk calls should be made more consistent with the national STD tariffs.
17. Greater consistency with the national STD tariffs would imply that any other changes in the these STD tariffs for the reasons mentioned above would also involve a change in the trunk call tariffs.
(d) Tariff rates for the non-peak-period
18. Any of the tariff changes arising for the reasons mentioned above will also affect the tariffs fixed for encouraging phone use during non-peak hours; peak time is 8:00 a.m. to 7:00 p.m. Tables 6 and 7, and Charts 7 and 8 show Indias non-peak hour tariffs which are provided for encouraging calls to be made in that time period . The tariff structure is devised so that,
Take for instance, calls covering a radial distance of 201 to 500 kilometers. The cost saving for the subscriber could be, for example, Rs. 22.50 per unit call of three minutes, if the call were made on weekdays between 7 to 8 a.m. or between 7 to 8:30 p.m. (see Table 7). In comparison, there could be an additional cost saving of Rs. 7.50 for such a call during early morning (6 to 7 p.m.) or late evening (8:30 to 11:00 p.m.); and if the call were made during late night then the additional saving might be only Rs. 3.75 per unit call.
19. The decision on which off-peak time to choose for making a call that would otherwise have been made during peak hour, is likely to depend on the comparison of the saving in cost and the additional inconvenience of making the call let us refer to this as the "net incentive". The current price structure suggests a relatively low net incentive for calls made during the late night period. Even among the other two alternative off-peak periods, the saving in cost is large in the period adjacent to the peak hour, and the inconvenience might be relatively much lower (i.e. there might be a higher net incentive for the off-peak period adjacent to the peak period).
20. It seems that further assessment might be needed to consider whether the incentives provided by the current structure of concessional tariffs result in greater likelihood of causing congestion in the non-peak period adjacent to the peak-period, and whether a greater number of calls could be encouraged during the other off-peak periods. This analysis could also involve a consideration of the cost differentials during peak and non-peak periods.
21. Another off-peak tariff scheme relates to international calls (Table 10). In that scheme, lower rates are charged after 11 p.m. irrespective of the time-difference from the country concerned. There might be a need to consider this time difference and the pattern of calls to different countries to decide on the timing for international off-peak rates. Another relevant point is that if we compare the off-peak time periods for domestic and international calls, we see that a much simpler pattern with fewer divisions in the off-peak period is adopted for international calls. It is worth considering whether the number of different time bands for domestic off-peak periods should be reduced.
IV.3 Some Salient Features of Telecom in India
22. Any consideration of policy changes with regard to Indian telecom tariffs should bear in mind the following features of the Indian telecom scene.
(A) There is a substantial unsatisfied telecom demand
23. This unsatisfied demand arises for two reasons. One, congestion in the network results in those currently connected to the network not being able to make all the calls they wish to make. Two, there is excess demand of those who have not yet got linked to the network but wish to do so. This is reflected by the waiting-list for connections.
24. As a result of the first type of excess demand, subscribers are off the prevailing effective demand curve (Annex 2). As a result of the second type of excess demand, the effective demand will itself shift when there is an increase in capacity (Annex 2). These factors make it extremely difficult to determine the demand curve or demand elasticity. Thus, implementing the Ramsey rule, or a price based on demand, becomes very difficult in India.
(B) Teledensity in India is low, and demand is likely to increase significantly in the next decade
25. It is expected that during the next decade, the telephone network in India would more than double its capacity. Thus, a virtually new network is likely to come in place in India.
26. The large expected increase in the network implies that if cost-based pricing were to be used, forward-looking costs, as embodied in the long run incremental costs, might be particularly relevant to the Indian situation. Furthermore, since considerable new investment by new entrants would be made to provide the service as a whole, the concept of TSLRIC might be appropriate as a starting basis for determining cost-based tariffs.
(C) There is inadequate information on demand characteristics, including demand elasticity
27. Information on demand elasticity for different telecom services is meager, even more so for India. The large expected increase in demand also implies that it will be difficult to calculate the elasticities of demand for the different types of telecom services.
28. Estimates from studies in other countries suggest low demand elasticities, particularly for local calls and for access to the phone. In such a situation, a price increase would actually result in a rise in revenue, and not much fall in demand for the service. This is even more likely for India because of the excess demand and supply constraints that exists for phone services (see also Annex 2).
29. A qualification to the above-mentioned point has to be considered because of another feature of elasticity of demand. The responsiveness of demand (or elasticity of demand) to a change in price is greater if the initial price is at a relatively high level. In the Indian situation, such a response could be expected for the relatively high priced services, particularly from those subscribers who are currently making only a few of these high priced calls.
(D) Available data on costs is not detailed enough to provide adequate information on the cost of providing different types of services
30. This is particularly true for basic telecom. Somewhat better data is available for value-added services, but even for these services more detailed account-keeping would be required for considering the relevant details of the various types of costs.
31. This feature suggests that if cost-based pricing were to be used, reaching a conclusion on the basis of detailed cost data will take time. If it were decided that the tariff policy should be cost-based, one possibility would be to formulate tariffs on the basis of the available (though imperfect) information. In this regard, new entrants or international estimates could be alternative sources for obtaining the required information.
(E) The recent opening up of the telecom sector has restricted private operators to certain service segments in specific parts of the country
32. This implies that the pricing of telecom services has to be determined under the constraint that the limited segment that is open for private investment must provide reasonable returns to attract investment to that segment. Of course, for a complete picture of profits for these investors, it is necessary to also consider the charges incurred for interconnection.
IV.4 Some Issues Regarding Telecom Tariffs in India
(a) General issues
33. For India too, there are primarily three questions to address,
In the Indian context too, the various options would be similar to those given in Section III.4 above.
(b) Social objectives and prices
34. The social objectives might imply a low price for access or for calls made by certain users/geographical areas. Examples of low prices include the free calls provided per billing period, a subsidy to rental, lower rentals charged according to exchange capacity and for certain rural users, and lower tariffs for certain rural telecom users
(i) Free calls
35. Conceptually, the free calls could be considered as an additional rental subsidy. Certain Indian subscribers, however, do not make calls up-to the free limit (see Table 2 above). For these subscribers, the free calls could be considered as a "lifeline subsidy", where a small number of calls are provided for those who might need these calls for giving a few important messages.
36. Thus, one option is to continue providing the free calls, but the deficit arising due to these calls might have to be adjusted in the price of calls or rental. Another option is to not provide the free calls. In that event it would be necessary to decide whether there should a two-tier price, namely whether a specified number of initial calls should be charged a lower/higher price than subsequent calls.
37. It was mentioned earlier that rental in India is charged on the basis of exchange capacity; reduced rental is charged for lower capacity exchanges. The lower capacity exchanges in India are a rough proxy for less developed telecom infrastructure and for rural and/or small localities. The relatively lower rental for such places is an attempt to encourage an increase in teledensity and to try promote the objectives of fairness and equity.
38. The following points need to be considered regarding different rentals in India being charged on the basis of exchange capacity. One, should the rentals continue to be lower for subscribers covered by exchanges with small capacity? Two, should the rentals be increased: for all subscribers, for a sub-group of subscribers, or not at all? If rentals should increase for a sub-group of subscribers, which criteria should be used for this purpose?
39. Another question to consider would be whether or not to reduce the number of different categories of exchange capacity which are currently used for pricing rentals. One extreme option in this regard could be to separate the subscribers only on the basis of whether they are rural/remote area subscribers or those from other areas, i.e. not distinguish subscriber according to capacity of the exchange to which they are linked.
40. Though the current scheme of rentals is divided into rural and non-rural, the only difference in rental is for one category of exchanges, namely those with 100 to 999 lines (see Table 8 and Chart 11). A question in this regard would be whether the current preferences given to rural areas should continue. In this context, additional possibilities include a removal of the preferences to rural area, or alternatively to increase the extent of preferences provided to rural area.
41. Certain general questions on the issue of increase in rentals were raised in Section III.4. In addition to those questions, there is a need to also consider that if a flexible option combining different rates for rental and usage were to be offered in India, should the same options be provided to subscribers in rural/remote areas and other areas. If not, then what should be the nature of the difference in the options provided to these categories of subscribers.
(iii) Lower tariffs for rural users
42. The various questions that arise in this context are similar to those raised above in the context of rentals.
(c) Overall Tariff Policy Decisions
43. If it were decided that prices should be based on costs, then a substantial revision would be required in the present accounting procedures, and collection of data. The process of obtaining full data on the relevant costs will take considerable time. There are basically two options for the tariff policy in the interim period. One is to maintain the current tariff structure, and the other is to change it. Thus a useful exercise would be to consider,
back to METHODOLOGIES FOR DETERMINING TELECOM TARIFFS
forward to INTERCONNECTION CHARGES
return to india_gii's home page