The Telecom Regulatory Authority of India (TRAI) has recently issued guidelines on minimum contention ratios for internet service providers (ISPs) and operators. The move is, being opposed by stand-alone ISPs for whom the cost of delivery will go up. Operators, however, stand to gain as they own last mile and can offer competitive prices. Stand-alone ISPs share their views on this issue...
Is the current regulatory framework conducive to the growth of stand-alone ISPs?
David Appasamy: The current regulatory framework is certainly not conducive to the growth of stand-alone ISPs in India. This becomes even more evident when you consider that today we are adding more than 10 million mobile connections a month, whereas after 10 years, we have just about 10 million internet connections in the country. Let alone a conducive policy framework, what we actually have is policy failure.
The reasons for this failure are clear. While the government announced the ISP policy in November 1998 and fixed the licence fee at just Re 1, the policies that followed ensured that ISPs were not financially viable and could not invest in developing the market. In the beginning, this appeared to protect the incumbent, but the practice has continued even after the advent of private telecom companies.
The incumbent was not required to separate accounting for infrastructure services provided to ISPs from retail services provided to consumers. This allowed it to charge high fees for infrastructure services (such as bandwidth charges and port charges for dial-up services), while undercutting ISPs on the price of services offered to consumers. This resulted in the exit of many large companies such as Wipro, BPL, British Telecom and Pacific Internet from the ISP space.
The focus of policy seems to be on charging high licensing fees from telecom operators, and letting them take care of everything. This short-sighted approach has resulted in the completely imbalanced growth of data connectivity and internet in India. For example, universal access service licence holders had the freedom to provide domestic calls on the voice over internet protocol (VOIP) platform since January 2006. Since they did not, TRAI was prompted to open up the option to ISPs. But policy-makers have done nothing about working on this recommendation.
The conflicting questions of who is responsible for internet growth, and who makes the required policies, contribute to the problem. The Ministry for Telecommunications and IT comprises two departments – the Department of Telecommunications and the Department of Information Technology. The Department of Information Technology is responsible for internet penetration, but policies are decided by the Department of Telecommunications. This situation is obviously not likely to lead to a conducive policy framework for internet penetration and growth.
Rajesh Chharia: The regulator has announced a number of policies for stand-alone ISPs and integrated players. But policy-makers take a long time to take final decisions. This acts as a deterrent to internet, especially broadband, growth. Take the case of internet telephony. TRAI had submitted its recommendations in August 2008 but DoT has failed to announce an internet telephony policy till date. We have to understand that 65 per cent of India's population lives in the rural areas, and for them, voice is the key application. Internet telephony, which is far cheaper than mobile or fixed line services, has to be used to promote broadband in rural areas. The policy-maker has to come forward and announce the policy for ISPs. The same is the case with IPTV. While there is a clause regarding net worth in the policy for stand-alone ISPs, the same does not hold for cable operators though ISPs pay a certain portion of their aggregate gross revenue to the government.ISPs have also been asked to vacate the spectrum that was earlier allocated to them, as this will be auctioned at a higher price. How will ISPs be able to provide broadband at low costs after bidding for and acquiring spectrum at higher costs?
Brajesh C. Jain: There is no need for the regulatory framework to be conducive for stand-alone ISPs. ISPs need to find a way to move forward and are resilient enough to do business. However, the difficulty arises when the regulations become retro-effective to curb any new innovations/applications developed to suit customer needs, both on technical and price-competitive bases. It has been seen that while the service is allowed to grow in the initial phase, when it reaches some level of generating worthwhile business, suddenly regulation is thrown in. Take the example of virtual private networks (VPNs). When the internet was initially introduced, incumbent operators were giving an intercity capacity of 2 Mbps at a price of about Rs 2 million.Reliabilty (uptime) was low, time to restore was high and smaller capacities were not given. ISPs worked very hard to develop VPNs with small capacity options at a fraction of the cost and better reliability. But then suddenly it was announced that VPNs were not part of an ISP licence.This is akin to saying that SMS/ringtones, etc. are not part of the cellular licence.
Jasjit Sawhney: Though the initial years were very encouraging for the ISP industry, there have been several policy measures in the past two to three years that seem to have been designed to protect every single inch of turf for the incumbent, while oblivious to the numerous advantages of the growth of the internet landscape in India.
ISPs have been brought under the revenue share net a few years ago. The logical step should have been to allow unrestricted internet telephony since a service that did not attract a licence fee earlier was now being subjected to a fee. A policy in this regard has still not been implemented, in spite of clear recommendations by the regulator.
Unbundling of the local loop would have gone a long way in non-replicating existing infrastructure, ensuring faster rollouts and increasing the subscriber base of broadband customers. However, in spite falling abysmally short of the targeted numbers, this point is not even at the discussion stage. In a number of countries, broadband has grown by allowing service providers access to the existing copper, with the operators paying the incumbent for use of this infrastructure.
If accepted, how would TRAI's proposal to specify the contention ratio impact standalone ISPs?
Rajesh Chharia: TRAI has announced guidelines for the contention ratio of bandwidth. While maintaining high contention ratios, telecom operators had been providing broadband services at low prices. They will now have to either revise their prices or face the heat from clients on account of low speeds.
Brajesh C. Jain: The objective of providing better quality of service (QoS) is to be appreciated. However, the actual reasons for the same need to be analysed. It is to be seen at what price point is a specific service given. For browsing and initial internet users, it is necessary to offer the service cheap. Those who use the internet for commerce, business, etc., there is need for higher throughput. This is the sustainable model rather than imposing a contention ratio for all, which would result in ISPs carrying extra bandwidth that is not utilised and leads to an increase in costs.Integrated telecom operators who own internet capacity are in a position to crosssubsidise and appear to charge for internet service as part of a bundled service and even on a stand-alone basis. They can offer predatory prices as they own the last mile and international gateways.
There are some measures that can be taken to host content within the country. Facilitating additional data centres within the country will help increase domestic traffic and reduce international carriage, thus reducing the cost of transport. Also, it is important to see that restrictions at the National Internet Exchange of India (NIXI) interconnection are eased. At present, the NIXI is not interconnected, resulting in inefficiency of traffic flow. Also, content providers are not allowed to connect to the NIXI directly.
Jasjit Sawhney: It may impact regional ISPs as they would have difficulty in managing the ratio at all given points in time. As they add more customers, they will need to add more bandwidth, and since the number of units of bandwidth that they sell and buy would be different, they would face some short-term issues. While the policy aims to provide the right quality of service to the customer, there could be many "ifs" and "buts" around it.In a market largely dominated by integrated telecom players, what does the future hold for stand-alone ISPs?
David Appasamy: Any discussion with DoT is reduced to ISPs being charged for not having paid substantial licence fees, or for having an insignificant user base. ISPs, the department seems to conclude, are not important in the larger scheme of things. What the department and indeed the government need to realise is that the sorry state of internet penetration is a larger problem for the country. Much like telecom penetration was considered an index of a country's economic potential 15 years ago, today internet penetration determines the country's ability to grow and compete in an interconnected world.
However, telecom policies are oriented towards large telecom operators, as evidenced in the pricing of spectrum for WiMax services, which has also been conveniently clubbed with spectrum for mobile use. Worse, spectrum already allocated to ISPs and being used for wireless connectivity for years, is now being recalled. As a result, when Wi-Max services are rolled out, they will be under the purview of telecom operators, with ISPs being effectively shut out of service provisioning.
Rajesh Chharia: Stand-alone ISPs are successful if the Retail Minus theory is considered. Currently, integrated operators offering internet and broadband services market their products either at the same rate or at lower rates than those offered by ISPs. In the US, in spite of the presence of large telecom operators, stand-alone ISPs also operate in the market. Similarly, in the UK, the incumbent was divided into different units for fair competition and growth. If in India too the same is being considered, stand-alone ISPscan play a vital role in developing internet and broadband penetration.
Brajesh C. Jain: Stand-alone ISPs definitely have a future. Like Cisco has moved past large telecom equipment manufactures, ISPs will also be able to make their way. Google is another example of using the internet to emerge with a very large canvas. There is a need for innovation, new products like cloud computing and personalised/customised internet applications. Users are looking forward to moving from browsing to converged services, and this calls for strong technical knowledge and commitment. A few pockets of excellence will not help. ISPs have demonstrated such capabilities in the past and have the inclination to do so in the future as well. Without a large number of stand-alone ISPs, the internet will not become a mass utility.
Brajesh C. Jain: A wireless technology will surely help ISPs. The mobile virtual network operator concept will have to be extended. ISPs are ready to exploit new opportunities, and wireless falls within the ambit of ISPs.
Jasjit Sawhney: ISPs will have to create their geographical and service niches beyond just access. Relying on access alone would be a dangerous strategy without significant scale of services, and service providers will need to provide the latest value-added services to their customers, which would vary depending on whether they are home users or businesses, and within businesses, on their size and industry verticals.
How are stand-alone ISPs likely to be impacted by the launch of 3G and Wi-Max?
Rajesh Chharia: The launch of 3G services is getting constantly delayed. The real benefits of 3G will be realised if next-generation networks (NGNs) are adopted by the main players. This should be combined with liberal policy-making for all telecom players, including stand-alone ISPs, for launching NGN services.
As far as Wi-Max is concerned, ISPs do not favour any particular technology and are free to use any new technology. This is the key reason why even DoT has preferred broadband wireless access (BWA) spectrum to Wi-Max spectrum.But looking at customer premises equipment costs, I do not think this technology can move much forward in the near future.
Moreover, in the earlier BWA policy, spectrum could be used for data but now the term itself has been removed, implying that spectrum can be used for any service. Since voice is a premium service, telecom operators who do not get 3G spectrum in the auction will try to take BWA spectrum for voice applications. This implies that stand-alone ISPs will not be able to get any spectrum for providing internet services to rural/urban users on a cost-effective basis.
Jasjit Sawhney: The success and uptake of 3G services is not yet proven for any telecom market. There will be a set of consumers who will take up the service and some of them may not use wired broadband any more. It is more likely to impact service providers who cater to the home user and the low-end broadband market. The cost of service is also likely to be a lot higher, especially in the beginning, and there would possibly be sets of consumers for both types of service.What are some of the key issues faced by stand-alone ISPs in India?
David Appasamy: India is slipping on the broadband front because we do not have free and open competition for internet services. This has resulted in the rapid decline of ISPs with only one national ISP left. Wholesale bandwidth prices in the country are many times more than in markets like the US.
Can we afford this? What happened to the vision of the New Telecom Policy of 1998-99 of "internet for all"? Can a country that is acknowledged as an emerging economic superpower risk its future in this short-sighted manner? If this state of affairs is to be reversed, it can only be done with pragmatic measures taken in the national interest – by recognising that internet penetration is a fundamental measure of a nation's ability to compete in a globally networked economy, and by policies that balance the growth of internet and data penetration along with telecom growth, by encouraging ISPs. Countries such as France have become role models for all of Europe with policies that have encouraged broadband penetration and use. We could emulate such a model.
The answer is clear. We must act quickly and decisively in the national interest, to encourage internet penetration and use with an ecosystem that encourages its adoption. Perhaps it is time that the Prime Minister's Office and the Planning Commission intervened, much as they did in 1998 to open up the sector, to rectify the sorry state of internet penetration in India.
Rajesh Chharia: When the government wanted to increase internet usage, ISPs were given incentives like the Re 1 licence fee. But now they are not being treated fairly. Licence conditions have been changed frequently, and applications that actually increase broadband penetration have not been promoted. The wish-list of standalone ISPs for the growth of broadband in the country includes:Jasjit Sawhney: Other than the ones stated in my response to the first question, another issue is the cost of leased lines. While the cost of internet/international bandwidth has come down significantly over the past few years, the cost of local loop has changed only marginally. This is cost-prohibitive for business customers who would want to take more and dedicated bandwidth.