Barely a decade ago, India had more telephone lines as compared to China; India had six million connections whereas China had only five and half million. India has done well in increasing its connections to about 22 million today. China, in the meanwhile have crossed 120 million connections and is adding about 25 million connections every year. It has recognized that providing telecom and Internet connection (built on a telephone network) to most of its people is a must, if it has to become a superpower in coming years.
Internet today is not just another means of communication. Those who use Internet regularly, know that Internet is POWER. It gives a user not only all kinds of information, but also enables him/her to do things one could not even dream of till recently. The power of Internet may be gauged by the following examples:
Internet gives to its users so much competitive advantage that those without access will face significant disadvantages. In other words, access to the Internet can enable people in all kinds of ways including providing access to education, removing barriers of distance and remoteness, and enabling one to get all kinds of information and to close business deals. But at the same time, lack of Internet access would put a person at a tremendous disadvantage. Therefore, unless steps are taken to provide Telecom and Internet access widely, one would face accentuation of
India today has about 22 million telephones and less than 0.7 million Internet connections for its one billion people. It is obvious that India urgently needs at least 150 to 200 million telephone and Internet connections. In this paper we examine the bottlenecks in achieving this aim and suggest ways to overcome these bottlenecks. We start in section 2 by looking in detail at these bottlenecks and suggest solutions. In section 3 we look at some of the R&D efforts in the country, particularly the recent efforts of the TeNeT group at IIT Madras in developing technologies, which provide examples of what needs to be done. In section 4, we look at policy issues which need to be pursued towards this objective. Section 5 concludes the article.
2 The Bottlenecks and Solutions
In a country like USA, household monthly expenditure of US$30* on communications is affordable to well over 90% of its households. Thus a telecom and Internet operator can expect a minimum of $360 per year of revenue from each household. If the operator invests $1000 per line to set up the service, the $ 360 per year return would pay for the finance charges, operation, maintenance and obsolescence cost and may provide some profit. The fact that some households and most businesses will provide higher revenue would only help. Thus a $1000 investment per line is viable in the USA. Technologies are developed to provide the best service at this level of investment. As the services is affordable to well over 90% of households, lowering of per-line cost makes little sense and does not enable the operator to expand market.The focus of R&D is therefore rightly not on reducing per-line cost, but instead, on providing a larger basket of services at this fixed cost.
Developing countries like India, on the other hand, are very different. Assuming that a household is willing to spend about 7% of its total income on communications, Table 1 shows the percentage of households in India which would be able to afford telecom service at various levels of cost. Thus, a revenue of over $350 (around $30 per month) is affordable to less than 1.6% of the households. If 30% of the households are to be provided service, the expected revenue has to reduce to around $10 to $12 per month. Further, the revenue expected must drop to around $5 per month, so that almost 60% of the 180 million households could afford communications. Assuming again that 35% of initial investment is required as yearly revenue to pay for finance charges (which is as high as 15% in India), and for maintenance, operation and obsolescence, one needs to bring down the per-line network cost to about $ 300 to cater to 30% of Indian households and $ 200 per line to enable almost 50% of households to have telecom and Internet connections (as shown in Fig 1).
| Household Income (yearly) |
|
|
| $2500 - $5000 | ||
| $1000 - $2500 | ||
| $ 500 - $1000 |
Unfortunately, the current cost of providing a telephone line in India is close to $800. While such a cost is affordable to most in the West, it is affordable to only a few in India. It is imperative for India to cut the per-line network cost by a factor of 3 to 4. The focus of R&D in the West is not on this cost reduction, but to expand services at a fixed cost which is affordable to most of them. It becomes the task of Indian scientists and engineers to turn the technological developments upside down, innovate and bring down the cost to enable Indians at large to have telephone and Internet connections.
2.1.1 The Focus on Access Network
The telecom network today consists of a backbone network component and an access network component (referred to as access network or AN). The backbone network consists of large switches and routers and interconnection of these exchanges and routers, including inter-city connections and international connections. Fortunately, the cost of this part of the network is reducing rapidly over the last fifteen years, driven by the R&D efforts in the West to provide higher and higher bandwidth on the backbone. The per-line cost of fibre and fibre-based network is decreasing rapidly every year. We therefore can borrow and use such developments intelligently. One must remember that the capacity of the optical network is doubling every nine months and per bit cost of the network is falling very rapidly. One must be very careful to understand the slope of this downward-moving cost and the technology. While one wants to install fibre for today, tomorrow and even for fifteen years hence, the network of today should be built to take into account the bandwidth requirement of only the next few years and yet must be upgradable. A careful design of backbone network would imply that a backbone network, even though very powerful, would not cost more than US $ 100 per-line including the cost of infrastructure.
The Access Network consists of the connection from the exchange (or ISP routers) to homes and offices and would today include an access switch. It is this part of the network which is expensive today, amounting to about two-thirds of the per line cost. In urban areas the access network cost would exceed $ 500 per line, and would be even more in rural areas. Besides, it is the access part of the telecom network, which requires maximum service and contributes most to the operation cost. It is the access part of the network on which one needs to focus one's attention.
Internet access today for the most part uses a modem to connect a computer to a telephone line. One then dials an Internet Service Provider’s telephone number and gets connected to an ISP router, which connects one to the Internet as shown in Fig 2. This seemingly simple technique has a number of pitfalls.
Internet sessions last a long time, usually an hour or more. On the other hand, telephone calls last barely a few minutes. The telephone network is India, and other developing countries, is optimally designed assuming that a telephone is used on an average approximately only 10% of the time during busy hours. The existing telephone network cannot handle a much higher traffic level.
As more and more people have long Internet sessions, the telephone network will get congested and collapse.
How do we expect to provide Internet connectivity to "millions" in such a situation? The high investment required to install a telephone line and the problems of providing Internet service on such a telephone network make our task very difficult. What is the solution?
A key to solving this Internet tangle lies in understanding the differences between voice and Internet calls. For the duration of a voice call, the 64 kb/s bandwidth of one PSTN circuit is almost fully utilized as the conversation is continuous. Internet traffic is, in contrast, bursty. The host at the other end takes time to respond to a request. Once the response is received, the user takes time to examine the data on the screen before sending the next command. Thus, the circuit is idle much of the time. In fact, it is likely that on the average only 1-10% of the 64 kb/s bandwidth of a PSTN circuit connecting an ISP may actually be used during an Internet call.
The solution will be to find a way of not occupying the telecom network resources 100% of time for Internet access. If a way is found to statistically multiplex Internet traffic from several on-line users and then carry the traffic through the circuit switched network, the ubiquitous telephone network would be capable of handling long Internet sessions of a large number of users. The key is that such multiplexing be carried out in the network as close to the users as possible.
2.2.1 The
Future Access Network
A conceptual future access network is as shown in Fig.3. Tomorrow's subscriber unit (SU) would provide direct interface to a telephone (or fax machine, cordless telephone, speakerphone, pay phone) as well as to a PC (serial port or Ethernet port) and would enable simultaneous voice and Internet communications. The subscriber units would be connected to an Access Unit (AU) using wired, wireless, fibre, DSL on copper, coaxial cable or even power line communications. A typical AU would serve anywhere from 200 to 2000 subscribers. In urban areas, the AU would be deployed close to the subscribers, almost on street corners, making this last-hop link less than 800m. In rural areas, an AU could serve an area of 10 km radius and sometime even 25km radius. The Access unit separates voice traffic and Internet traffic. The voice traffic is switched to the telecom network (typically using an E1 connection to an exchange using the standard V5.2 access protocol). The bursty Internet traffic from each subscriber is statistically multiplexed before being carried to the Internet network. Ability to multiple bursty Internet traffic from multiple subscribers at AU enable one to overcome the bottleneck described earlier in the section.
Such a conceptual Access Network solves the Internet tangle as the Internet traffic does not enter the telephone network at all. The concentration of traffic at AU implies savings in backhaul bandwidth between AN and exchange/router. Besides, placing of an AU close to the subscriber (at least in dense urban areas) implies lower cost for the last-hop.
As mentioned earlier, it is the Access Network, a point to multi-point network, which is most difficult to maintain and requires significant servicing. Besides most telephone faults occur in the Access Network, and are responsible for large down-time of telephones in India.
| Self-employed/small business |
The Indian economy consists of two sectors, an organized sector, which probably provides 10% employment, and an unorganized/small business/self-employed sector. The employees in the former typically works 225 days a year, work for 6-7 hours a day as shown in Table 2 and still earn about three times that of those employed in the latter sector, where the employees work about 325 days in a year and for about 10-11 hours a day. The important thing is that for any service that works cost-effectively and provides service to a significant number of people in the country, requires the organised sector to use this unorganized sector in a judicious way. One cannot imagine newspapers and milk reaching each doorstep at an affordable cost without the unorganized sector. The Public Call Office, which today generates about 20% of total telecom revenue in some states, and enables large sections of people in India to use telephones, provides an excellent example of how the telecom department has used the unorganized sector in expanding service. If the cable TV reaches over 35 million homes, today it is due to the cable TV operators. The cable TV operator stays just down the road and carries out repair work almost immediately, even on weekends.
If the telecom and Internet is to expand to provide service to hundreds of millions of people, it is imperative that a judicious combination of the organized and unorganized sectors is used. This implies that the Access network, the portion of the network which require most efforts in servicing, be franchised. A nation-wide or a state-wide operator could operate the backbone network; the access port on the other hand, could be franchised to an operator providing access service to a community. This implies that provision and maintenance of the last mile (including Access switching), finding the subscriber and installation of subscriber premise equipment, bill collection – all these became the tasks of the franchise operator who runs it as a small business. This would not only significantly reduce the cost of operation, but also improve the service provided to a subscriber. Of course, Access Network (AN) technology should permit this. At the same time, it is highly desirable that the AN technology should be such as to enable more than one franchise operator to operate in one area, giving choice to a customer.
With the current costs of Telecom Network in India hovering around RS 32,000 ($800) per line, reducing cost to around Rs.10, 000 ($250) per line is indeed a difficult task. But then R&D tasks are not supposed to be easy; and a vision of enabling India through Telecom and Internet cannot be a simple task.
The task may be difficult, but is achievable. This confidence emerged in the late eighties primarily due the to efforts of the Center for Development of Telecom (CDOT), a telecom R&D organization set up by the Department of Telecommunications in the mid-eighties under the charismatic leadership of, Sam Pitroda. CDOT developed telecom switches and these switches today provide 12 million of the total of 24 millions telephones in India. The switches are rugged, work without air-conditioning and even today cost 2/3 of any imported switch.. CDOT has continued to upgrade the switch and have added ISDN services, signalling system 7 (SS7), access protocols like V5.2 and even Intelligent Network (IN) functions.
But more important than the product itself, CDOT generated a confidence amongst Indian engineers that world-class design efforts can be undertaken in India. The large number of people that it trained, provided the basis for numerous design-houses that emerged all over the country.
About five years back, a university-based group, called Telecommunications and Computer Network (TeNeT) emerged at IIT Madras, Chennai. This group focussed on developing a world-class and yet affordable Access Network for India. It helped incubate a number of R&D companies formed by its alumini, Midas Communications, Banyan Networks, Vembu Systems and Nilgiri Networks, thus formed, started working in partnership with IIT Madras to develop a wide range of access technologies. The products designed by this group today include Fibre Access Network (optiMA), Wireless in Local Loop system (corDECT) and a Direct Internet Access System (DIAS). The products are aimed at significant reduction in access cost and at the same time enable large scale usage of Internet. These products take into account that the future access network should separate the telecom and Internet traffic at the Access Unit (as shown in Fig.3), and that the connectivity from the AU to subscriber could use any of the several media. Let us take a brief look at these products
3.1
corDECT WLL
The first access product developed is corDECT WLL, shown in Fig.4. It consists of a subscriber unit called Wallset with Internet port (WS-IP), located at subscriber’s premises. It has a standard (RJ-11) interface for a telephone and a serial port (RS 232) to connect a PC (without use of a telephone modem). The system provides Internet at 35/70 kbps and simultaneous telephone conversation. The WS-IP is connected to a Base Station (CBS) on wireless and the base station is connected to a Access Unit which consists of a DECT Interface UNIT (DIU) and an iKon Remote Access Switch. The AU separates the voice traffic and switches it to the circuit-switched telecom network using E1 interface with either V5.2 or R2-MF signalling. The bursty Internet traffic from multiple subscribers is statistically multiplexed and connected to an Internet Router using one or more E1 interfaces. A typical AU serves 200 to 1000 subscribrers. Fig 5 (a) and (b) show the pictures WS-IP, CBS and the DIU. The wallset-base station wireless link distance could be as large as 10 kms (using line of sight connection) theough it is likely to be 1 to 2 kms in urban areas. To serve sparse rural areas a Relay Base Station (RBS) could be installed on a tower upto 25 kms away from a CBS and in turn serve subscribers in a 10 km radius using a two-hop DECT link.
The cost of an installed corDECT line in India varies from Rs.14,000 ($325) to Rs.18,000 ($425) depending upon the deployment scenario. A multiple subscriber unit, referred to as Multiwallset (MWS) has been designed to serve four independent subscribers in the same building. The MWS brings down the per-line cost of corDECT by half.
3.2
DIAS
The second access product, again modeled after the Access Unit of Fig.3, uses DSL-on-copper technology to provide access. As shown in Fig 6, the Direct Internet Access System (DIAS) consist of a Basic Rate Digital Subscriber Unit (BDSU) or High–bit rate Digital Subscriber Unit (HDSU) at subscriber’s premises. The BDSU connect a telephone and provide an Ethernet interface for one or more computers. It is connected to an Access Unit referred to as Internet Access Unit (IAN) using twisted pair copper wires. Typical distance is 800m, though 5km it is possible to use it a distance using 0.4 mm twisted copper pair. The service provided is 144 kbps always-on Internet Access the rate of which drops down to 80 kbps when the telephone is being used. The IAN separates voice traffic and directs it to the telecom network and the Internet traffic is taken to the Internet Router. HDSU, designed for corporate subscribers, provide upto 8 phones and an Ethernet connection for Internet. Upto 2 Mbps always-on internet connectivity is possible with the copper distance being restricted to 3 kms.
The BDSU-IAN cost would be around Rs.12,000 ($275) per subscriber not including the copper cost (which would be about Rs.5,000 or $125 for a 800m connection). The HDSU cost would be almost double of this. Fig 7 shows a picture of a IAN and a BDSU.
3.3 Access Center
The third Access product designed by the TeNeT group is a simple POTS line connected to an Access Unit called Versatile Remote Unit (VRU) located on a street corner. The key is that it costs only about Rs.2,000 ($50) excluding the copper cost and provide amongst the cheapest way of providing telephones.
The interesting thing is that the corDECT WLL, DIAS and VRU can be combined into a single Access Centre as shown in Fig 8. An Access centre could be located on a street corner in dense urban areas serving 200 to 2000 subscribers in about 800m radius. Alternatively, it could be located in a small town/village centre and serve subscribers in 10km radius or even 25km radius.
The corporate and upper middle-class subscribers are served using DSL and provided withalways-on Internet connection. On the other hand, middle-class subscribers are provided telecom and Internet using wireless; and the lower middle-class subscribers could be provided service using either corDECT multiwallset or POTS (using VRU). The AC can be made self-contained with a built-in power-plant and battery back-up as shown in Fig.9. The ratio of DSL, WLL and POTS subscribers could be varied depending upon the locality served and the subscriber profile.
The Access Centre provides one of the most versatile ways of providing telecom and Internet access. Access using coaxial cable or even power-line can be added to the AC tomorrow.
A Fibre-in-the-Loop system is yet another product from TeNeT group. A fibre drop-and-insert backhaul system connects multiple ACs, (discussed in the previous section and located on street corners), together using a PDH or a SDH ring network as shown in Fig.10. The Voice and Internet traffic from each AC are taken to a central location in the city and handed over to the telecom switches and Internet Routers respectively. Alternatively, if the AC is located in remote locations, where fibre may not be available, a point-to-point microwave radio (2GHz/ 7GHz / 11GHz / 13GHz or 17GHz) can be used to bring both the voice and Internet traffic to a town as shown in Fig.11
3.5 Network Management System (NMS)
Finally, the TeNeT group has come up with an integrated network management system (CygNeT) to manage all these products as well as other telecom and network products. Integrating SNMP and TMN, the CygNeT manages traffic, configuration and health and enable a multi-tier operation. Fig.12 shows a typical view obtained from CygNeT.
The network described above is versatile, flexible and expandable. It provides different kinds of services including voice telephony, dial-up Internet connections and permanent Internet connections. The network can be deployed in large cities, small towns as well as in sparse rural areas (using corDECT Relay Base Station).
Fig.13 shows the typical deployment plan in a large urban centre of Chennai. Access centre locations are identified at about 20 places, each serving roughly an area of 1.6 x 1.6 km. About 1000 subscribers are provided connectivity from each access centre. To get the service quickly off the ground, first a few ACs can be connected to the switching centre using point-to-point 8 Mbps radio links. In the mean-time, the work on a SDH fibre-optic SDH ring network can start to connect all the ACs to the switching centre. What is important to note that this network not only provides voice connectivity but also Internet connection without loading the telephone network. This is because, the Internet traffic is seperated at Access Centres and carried on from this point on a packet network. Fig.14 shows similar plans for installation of the system in Hyderabad.
Fig.15 shows a plan of the network designed for a rural area. Thanjavur is one of the better-off districts of TamilNadu with rich agricultural land. The proposed network consists of two Switching Centres located at Thanjavur and Kumbakonam towns. Several Access Centres are located at small towns all over the district. The AC serves not only the town, but also the rural area around it. By using Relay Base Stations, a subscriber can be provided both voice as well as Internet connection as far as 25 kms from the town. The ACs are connected to the Switching Centres using point-to-point microwave radio link.
Fig.16 shows deployment of corDECT system in Connaught place in Delhi, one of the Delhi’s highly built-up business districts. Using four BS, telecom and Internet Access is provided in the whole area by an ISP
Recently DOT installed a pilot corDECT system in Kuppam town in Chittoor District of Andhra Pradesh (shown in Fig.17). With BS mounted on two towers, telecom and Internet is being provided in about 50 villages (some of which are shown in Fig.18).
The TeNeT group continues its innovations in adding features and services to its Access Network while bringing the cost down. Both corDECT WS as well as DSU of DIAS system has been designed with a built-in DSP with spare memory and computing power. Voice on Internet can thus be added to these systems when it is legally allowed and when QOS provisions on Internet enable its proper use. The connection from the AU to the PSTN will then become redundant and could be removed. The TeNeT group is examining the possibility of developing a cable access system providing voice, video and Internet, which could work in the adverse cable environment of India. The wireless connectivity on corDECT is being enhanced to provide 384 kbps packet-switched service on corDECT by the end of the year and multilevel transmission is being examined to enhance the rate to 2.3 Mbps by the next year. 3G radio interfaces are also being examined for use in the coming years. The DSL is being enhanced by addition of ADSL and VDSL. Finally, the backbone interfaces of the AC are being enhanced to incorporate QOS.
In summary, one longs that using the technologies developed by the TeNeT group, it is possible to set up a telecom network today at a total cost between Rs.17,000 ($400) to Rs.20,000 ($475) per line.
4.1 Implications of 150 to 200 million telecom and Internet connection target
A goal of providing 150 million to 200 million telephone and Internet connections is not just a noble goal, but has several implications. The large target implies that Telecom and IT is no longer confined to large cities or amongst wealthier sections of the society, but reach each small town and rural area and to all sections of people. This could enable rural youth to receive better education and train themselves to stand up in this fast changing world. It could result in software companies being set up in smaller towns and rural areas, initially as satellites to large software companies in the city (just like the Indian companies in early 80’s attached themselves to software companies abroad), but slowly coming in their own. It could enable someone in a small town to set up a design house, carrying out most complex designs of virtually anything, a task which was hitherto confined only to a few metros. But above all, this could create a confidence amongst the youth in small towns and rural areas — the kind of confidence to deal with the world as equals, that we have started seeing in recent years in large cities of India. In fact this goal of 150 to 200 million telephone and Internet connections opens up a possibility of comprehensive social development. But this does not imply that setting up Internet connections itself would automatically lead to all-round social development. These are mere possibilities and active intervention would be required on various fronts to achieve this. The telecom and Internet connections, however, would open up the possibilities as a potent enabler.
Another implication of the goal of having 150-200 million telephone and Internet connection is that India would have to manufacture such a large quantity of equipment, install and commission the lines and maintain them. The numbers being very large, this in itself would create an economic activity in the country which would be unparalleled. Even if we assume that Rs.2500 ($60) per year is required to service a connection, 200 million connections imply Rs.500 billion (Rs.50,000 crores or $12 billion) of economic activity every year. This is by no means a mean economic program.
Yet another implication emerges from the affordability issue discussed earlier, namely that the cost of a telecom line needs to be reduced by a factor of 3 to 4. An effort to reduce cost by a factor of 3 to 4 of the internationally prevalent cost would propel the R&D in this sector to be the very best in the world. This in fact therefore is not merely a program to expand telephony and Internet; it is in fact a program to become a Technology Leader in world in this sector.
Finally, a market size of 150 to 200 million (coupled with at least as large a market in other developing countries facing similar problems) is not small. Even at a cost of Rs.10,000 ($250) per line, 200 million lines implies a market size of Rs.2000 billion (Rs.200,000 Crore or $50 billion). The market size is comparable to the telecom market in the west. There is entry reason for industry to view this opportunity with great interest.
4.2 India has the capability
As discussed in Section 3. India has the capability to take up this challenging task of significant cost-reduction. The efforts of CDOT and that of TeNeT group (at IIT Madras), described earlier, indicates this. But even more important, India today has a large number of telecom and IT companies, which carry out major design and development tasks for the best products in the world. Bangalore, Chennai, Hyderabad, Pune, Mumbai and several other cities are full of such companies. The problem is that most of these companies are carrying out the development tasks today as service work for industries in the west. Not only are the products owned by companies in the West, but also the products are designed to primarily serve the booming market of the West. The Indian companies are definitely capable of shouldering these challenging tasks, and without the efforts from many of them, no one CDOT or TeNeT group is capable of developing what is required for 150 to 200 million telecom and Internet connections.
The question is: why do these companies not work for the Indian market? The reason is that they do not have the confidence that this potential Indian market would really materialise. They do not believe that the Government policies would allow opening up this market. They are afraid that beuracratic decision-making in India will not let them benefit from such a potential. They would instead continue the service to the multinationals, where their returns are assured.
4.2 Policy is key
Therefore a sound telecom policy is the key to generate such confidence in the Indian R&D community. The government not only has to come up with policies which enable Indian R&D to contribute towards India, but also has to be seen to promote this. Indian telecom policies have been singularly lacking in this regard.
Even though DOT started privatisation of telecom operations in 1994, with a view to liberalise the environment so that multiple players can use different technologies to help expand telecom network, the process has been bogged down. Though the National Telecom Policy 99 attempts to untangle the web, the policy does not display a basic understanding of the affordability issues. That significant reduction of cost is imperative and that access needs to be franchised to the self-employed unorganised sector does not appear to be a consideration, while formulating these policies.
The privatisation process of telecom allowed prospective operators to bid for the right to operate in a circle (whole state). States in India are large and several thousand crores of rupees (several hundred millions of dollars) are required to attempt to build a telecom network for a state. Of course, only large corporate houses in India and multinationals could take up such a task. Furthermore, without an understanding of the paying ability of the large Indian population, these corporates bid very large amounts to obtain the license. Unable to pay these amounts and faced with bankruptcy, they prevailed upon the government to change the license fee payment to revenue sharing in NTP 99. The NTP99, formulated under this environment, did the best that was possible.
But an important question is; why should privatisation start with a state-wide operation a all? Why cannot the Access part of the network be franchised or privatised? This would have enabled tens of thousands of small operators to come up each with modest investment to provide services in a community (small area), much like the cable TV operations today. Why has privatisation process not proceeded on this line? If one also leaves it to the franchise operator to chose whatever technology he/she wishes to and only work out a revenue sharing and interconnect agreement, it would really unshackle the communications field. An R&D company, coming up with an innovative product, has to get just a few of the thousands of franchise operators to try out the product and not depend upon its acceptance by beauracrats, or Basic Services Operators with multinational tie-ups. This single approach could get many of the Indian R&D companies to start looking inwards.
Another problem that bogs down the R&D efforts towards developing products for the Indian market, is that the Indian design houses are today not big enough to take on the might of the Motorolas and Alacatels of the world. This does not mean that their technology is not good enough; in fact a reasonable part of the technologies of these multinational telecom giants is now being created in India. What they can not match is the financial muscle of the multinationals. Statewide operators need large financing to carry out the operation and vendor financing, where system vendors not only supply the equipments, but also provide financing for these equipments. This is key to the purchase decision. Most multinationals are able to arrange financing from Exim banks of their countries. The financing terms are extremely attractive – an interest rate corresponding to LIBOR + 0.5%. (nearly 6% rate) plus five years moratorium on payments. Operators are cash-starved in the first few years and the five year moratorium is extremely handy. Coupled with attractive interest rate (compared to about 15% interest rate within India with no moratorium on payments), the financing terms are such that an operator is ready to buy an imported product at 100% higher price as compared to a comparable indigenous product. Such a situation is a big barrier to Indian design houses designing for India.
Ultimately the country loses on two counts. First, India uses a high-priced product, for which payment has to be ultimately made; and the higher cost of the telecom network implies that telecom remains confined to a small percentage of India’s population. Secondly, the Indian design house is discouraged from designing for India. Undoubtedly, the Indian policy makers have to pay serious attention to overcome this situation by coming with attractive financing policies for indigenously designed products. But what would help immensely is if they adopt the privatisation through franchising policy. This would create thousands of small operators, who cannot be all financed by multinational vendors. These small operators depend much more on their own resources and try to minimise their costs. They are much more likely to adopt the indigenous lower-cost products, to the ultimate benefit of the consumer.
4.3 Other Policy Initiatives
Another policy initiative, which is a must, is an active encouragement for setting up telecom and Internet service in small towns and rural areas. A simple solution is to provide a higher revenue share (for the access franchisee) for operations in small towns and rural areas. Similarly, in order to encourage expansion of telecom and network to lower middle classes and poorer sections in the city, a higher revenue share can be provided to a franchise operator who provides connections in slums and houses of lower income groups.
It is not just enough to encourage indigenous R&D, but encouragement to indigenous manufacturing is equally necessary. It is simply not possible to import $50 billion worth of equipments. Indian manufacturing has to be strengthened and made world class if we have to achieve our target. Clear policy initiatives encouraging indigenous manufacturing is required. For example, sales tax and excise duties have to be reduced. But at the same time, we have to push our manufacturing houses to produce at world class standards at a cost comparable to that in the rest of the world. China, for example, has left us far behind in this sector and urgent action is required to rectify this situation.
Unfortunately with the budget deficit of the Indian Government reaching a mammoth amount, sectors like telecom (and telecom manufacturing) have been seen as a cash cow, and high taxes have been imposed on this sector. This itself increases the cost of telecom and makes it unaffordable to most of its population. An alternative would have been to encourage telecom growth by lower tax rates — the resulting larger volume would not only compensate for the lower tax rate, but also propel more activity in this sector, bringing in indirect revenue to the Government. Even though the large budget deficit clouds thinking in this direction, efforts have to be made to break this shackle.
To conclude, providing 150 to 200 million telecom and Internet connections could enable millions of people in India and transform India. The most important resource in this century is not going to be material resources, but human resources. The expansion of the telecom and Internet network is the key if one has to convert India’s large population into a large pool of human resources.
The expansion of the network however is not possible at the price point prevalent in the West. Significant cost reduction is required and the efforts of CDOT and TeNeT groups points out that this is indeed possible. But one need to tap the large number of Indian electronics design houses to make this possible. Appropriate policies are necessary to make this possible. Opening up franchised telecom operation in small localities could be an important step in promoting such an effort.
Yet, all this would be only a beginning. The article has not dealt with several other concurrent actions that are necessary. First of all, the Personal Computer, as it exists today is too costly and inappropriate to be used by hundreds of millions of people in India. Efforts to come up with a variety of low-cost access terminals and integrating these with communication devices is a must. Secondly, Internet Access can not reach millions without incorporating local language and local content in a massive way. Efforts so far in this direction have been rather poor. Similarly expecting everyone to use keyboard to access all information may limit the use of Internet. Voice-enabled Internet service (with Indian languages) must emerge soon. Finally, Internet in India requires Indian applications – applications that emerge from the Indian way of life.
In spite of all the massive efforts required, this goal is very much realisable. The very enunciation of this makes it appear to be very feasible. Only this optimism need to be converted into action.
Author
Dr.Ashok Jhunjhunwala
Ashok Jhunjhunwala (M '84) received his B.Tech. from IIT Kanpur, India, in 1975, and M.S. and Ph.D. from University of Maine, USA, in 1977 and 1979 respectively, all in Electrical Engineering. During 1979-81, he was on the faculty at Washington State University, and from 1981 he has been at IIT Madras. He is a Professor and currently the Chairperson of the Electrical Engineering Department.
He is a co-founder of the TeNeT Group at IIT Madras. His research interests are communication systems, computer networks, fibre optics, and surface acoustic wave devices.