India had 254.40 million internet subscribers by September 2014. That’s an internet penetration of 20.39 per 100. But the picture changes when you consider that 70.23% of them are narrowband subscribers and only 29.77% access a useful connection. Less than 6% of total internet subscribers use what TRAI defines as ‘desired bandwidth speeds’. Surely digital inclusion is about more than such basic connectivity? And while government policies such as Digital India set targets for the supply/infrastructure side of digital inclusion, who is looking at adoption or demand-side issues?
Digital India has become the centrepiece of the current Union government’s plans. It is expected to be a key driver of growth and governance, and Digital India’s stated mission is to “transform India into a digitallyempowered society and knowledge economy” (1).Internet connectivity is key to such a digitally empowered society. This piece describes what connectivity means and what the different aspects of this connectivity are. It also looks at the present status of connectivity based on TRAI data, and examines attempts being made as part of Digital India to ensure connectivity. Lastly, it shares some concerns about the road ahead if the country is to realise the full potential of such connectivity.
Internet or World Wide Web, which gives the now not-much-in-use www as part of the address used in browsers is the interconnection of billions of things across the world. The reason the word ‘things’ is used instead of ‘computers’ or ‘devices’ is because,over the last 20 years,the objects being connected have moved from computers and devices to refrigerators and cattle(2). As a worldwide network, it connects all thosewho want to be connected. The connectivity of cattle obviously does not imply their agency, but that of the ranchers. Be that as it may, between want and need hangs ability – ability that is not merely technological or economic.
Technologically, if there is a pipe that connects all computers, then we would say there is potentially a network that anyone can tap into. This means wiring up all computers and all other objects that can connect to this network. Imagine then, an ATM in a village. To work as an ATM, it needs to be wired to the banking system’s network, which will allow its seamless use by those who have access to a debit card. Imagine a dispensary in an urban hospital where the stock of a medicine can be accessed by a warehouse and replenishments sent when the stock of say, paracetamol, reaches a prescribed limit. Now imagine how dispensing cash or replenishing medicine would work in a world that’s not wired. It would mean requisition slips, endless phone calls and memos to a warehouse. What the connectivity does is reduce delays or inefficiency in the system. Note that while the final beneficiary of these examples may be the citizen, the technological connections that enable it are still between bank and ATM or dispensary and warehouse.
This vast interconnectedness of all potential objects to each other, the efficiencies it entails, and the jobs that the creation of this infrastructure would bring, is but one part of the phenomenon that we call net connectedness. The creation of jobs and businesses around the physical creation of this network is an important reason why net connectivity is touted as an economic force multiplier. In fact, the reduction of barriers in doing business, and the creation of new businesses and opportunities around the use of this internet, has led to the net industry being treated as a general purpose technology, which implies that, like the automobile or steam-engine,it changes the very way in which society is configured and lived (3). What is important in this understanding of the net is that it is a wire or a pipe that connects (the connectivity that is called ‘internet’ consists of wires and air – spectrum — and may involve telephone, cable, mobile, satellite and other wireless modes of connection; the word ‘wire’ or ‘pipe’ is used as a metaphor.)
There is a connected, albeit analytically different way of construing the internet. Internet as a means of conducting transactions – personal transactions (emails and social connectivity); commercial transactions (buying and selling of goods and services); managerial transactions (managing businesses, governments and other entities) and democratic transactions (partaking of the benefits of, and participating in, governance). In passing, it may be mentioned that the internet as a source of knowledge belongs to the world of transactions because there is a seeking and receiving of knowledge and information qua service. The difference between the internet as a pipe and the internet as a means of conducting transactions is that the focus changes from the carrier/medium to the content that passes through, content that enables citizens or organisations to potentially conduct transactions. The moment the focus changes to the content that passes through the pipes, we enter into the human capabilities that would be needed to conduct the transactions. The potential benefit that this connectivity provides to such connected individuals is deemed important enough, and its absence inequitable enough, to demand and insist on the inclusion of every citizen in this network.
In order to appreciate what inclusion involves, it is necessary that we understand what this technology entails in its operation. For any two computers (or objects) to be connected there should be a way in which these two can talk to each other. Traditionally, the only way devices could talk was over a wire. Recall the old phone lines. Later wireless connections became possible. Despite all the improvements in technology, even today the amount and speed of talking that can be done over wires is far more than can be done using the airwaves. In fact, we need big pipes (in terms of capacity not necessarily in diameter and radius) to ensure the connections across the world. Imagine a computer in rural Africa trying to speak to a computer in rural India. In its simplest configuration, there would be a wire that connects A, our African computer, to I, our Indian computer. But since that is not a financially feasible way of going about it, there would be a network of wires. So a wire from A’s home to a junction (quite like a village road), where it joins the wires that reach the highway (quite like the state highways), and the wire that travels all the way to a port city (quite like the national highway) where there is a wire that goes under the sea to the Indian port city. And then the same way the wire reaches I’s home. This is exactly how the internet is formed and connected.
This example brings home the enormous length of wire needed to ensure internet connectivity across the world. To make it easy to distinguish between the different kinds of wires used between the home to first junction, first junction to statehighway, national highway to port city and port city to the other port city, the telecom nomenclature of last mile, middle mile, national backbone and under-sea cables are available to us.(4)
It is clear from this that the physical aspect of the internet is not easy to lay, in terms of both cost and time. Since the people likely to use this infrastructure and pay for it are likely to be in the cities and wealthy, it is obvious that these pipes are likely to be found in richer countries and richer cities; and the onus for laying these pipes will have to lie with the government, especially in those countries where the income levels are not high enough for citizens to pay for this infrastructure directly.(5) India is no exception in this regard, as we will see later.
But assuming that there is a pipe that comes right to our homes, we have four related issues to contend with. Does an individual need the content that can be delivered through this pipe? Does she have the ability to utilise the content delivered? The ability to interact over the pipe?And the ability to pay for the content being received?
To begin with the last question: there are two financial costs to be paid for accessing the material — the cost of the device and the cost of connecting the device to the pipe. Unlike say electricity or gas or water which require a relatively rudimentary device of comparatively lower cost to utilise what comes through those pipes, the device that is needed to utilise what comes over the internet is still expensive, and the more expensive the device, the better the interaction with what is out there. Assuming that this one-time asset cost is paid and a good device procured, there are two other costs. Electricity costs to run the device (even charging the phone has a cost) and access payments that allow the use of the pipe. Both are variable costs and, in the case of access, can be substantial (relative to income) if one considers how much use (bandwidth and download/upload) is likely to be made of the pipe. So cost is an important consideration for the use of the internet. Though related, it should also be pointed out that electricity is an important issue, because there is no guarantee that electricity is available all the time. So even if paying electricity cost is not a worry, there is a supply constraint that affects internet access.
Then there is the question of whether a person really needs the material/content that can be delivered over the internet. There are again two aspects to this. One can argue that there may be some who do not have any use of this material, and this may not be entirely inaccurate. But there is another related point, which is that people are not even aware of what is available online to realise whether any of it is of use. While the first issue is of interest, the second is of awareness. They do not know what this pipe can bring or take or do. But this lack of awareness is not the same as lack of competence.
If there is awareness, and individuals are aware that the pipe is of potential use, then too they may not have the necessary competence to utilise it. And once again, there is not just one set of competencies. The basic competence that may be lacking is the ability to use the device. They don’t know how to switch on the computer and operate it. They don’t know what the icons do and how to use a mouse. These hurdles can be labelled computer illiteracy. But there is another incompetence, which is related to the third of our four questions regarding the use of the internet – the ability to interact using the pipe. This competence goes beyond the ability to operate the computer or device and is related to understanding what the vast network of computers has to offer and how to search and locate things of interest – in short, how to interact.
Now this competence is closely linked with at least two barriers – linguistic and technological. Linguistic competence is crucial because the vast majority of things on the internet are in English and certainly not in Indian languages.(6) Even if the material is there, there is a technological barrier because there is no ease of use from keyboards to fonts. When you add the barriers that those with physical disability involving sight or sound may have, or even those who are not comfortable with typing on a keyboard, or those who are old and hence find learning basic technology difficult, we have another set of barriers to the use of the internet. Compounding all this is the lack of basic literacy, which can make all these even more difficult.(7)
Now, when we take all these barriers arising from the two basic technology and non-technology barriers, often called supply-side and demand-side barriers or access and adoption barriers, then we are facing the sum total of all that constitutes the inequalities that divide that part of the population that can be said to be on the net and those outside its pale. And if we believe that having such net connectivity leads to a better life,(8) then exclusion from this network on any ground leaves the person in a position of disadvantage and inequality. Hence, internet penetration or digital literacy that merely looks at ability to operate computers or internet literacy that merely shows how to connect to the internet and open a browser, are not adequate to understanding what is at stake here.For digital inclusion there is the entire panoply of services and competencies that need to be available. So a purely technological access does not translate into digital inclusion because the inclusion needs to be meaningful.
Current status of digital inclusion
The preceding discussion should make it clear that the length of wires or pipes is just one measure of digital inclusion. The use of these pipes or connectivity as indicated by the actual number of internet connections, whether broadband or narrowband (512 kbps and 512kbps respectively), gives us an idea of the number of people accessing the internet. Recently, the Telecom Regulatory Authority of India (TRAI) has come out with the telecom performance indicator report for June to September 2014.(9)
Table 1. Internet subscribers at a glance, TRAI, Sept 2014
|Total internet subscribers|
|Wired internet subscribers|
|Wireless internet subscribers|
|Total internet subscribers per 100 population|
As Table 1 shows, there are 254.40 million internet subscribers. But the vast majority, 70.23%, are narrowband subscribers. If we recognise that narrowband connections are increasingly insufficient for full use of the internet, we have a rather dismal 29.77% of internet subscribers who access a useful connection. If we look at the split between wired and wireless subscribers, then we have just over a shade more than 7% of subscribers using a wired connection. Now consider what the TRAI Consultation Paper on Broadband has to say:(10)
‘For all practical purposes the DSL, Cable Modem and Fibre to the Home account for nearly all broadband subscriptions in the country. Mobile Wireless Access Networks are the other main source of delivering broadband. However, due to the paucity of spectrum, they are not equipped to provide the desired bandwidth speeds.’
If TRAI’s own benchmark is used to read the above data, it becomes clear that with 15.13 million subscribers of wired broadband (See Table 2 below), less than 6% (5.95%) of total internet subscribers are using ‘desired bandwidth speeds’.
Table 2. Trend of internet subscriber base (in millions), TRAI, Sept 2014
Mode of Access
Fixed Wireless (Wi-Fi, Wi-Max, Radio & VSAT)
Mobile Wireless (Phone + Dongle)
While India’s internet subscriber penetration rate per 100 is 20.39, if we calculate the subscriber per 100 rate for those using ‘desired bandwidth speeds’, then the number falls below 2 per 100 using 2011 census data. So what we have effectively is a miniscule population of the country on the internet in any effective sense, if by effective we imply a substantial use of what the internet has to offer.
Some sense of what I am calling this miniscule effective presence on the internet can be seen if we compare the penetration of telephones with that of the internet. The teledensity for the country stands at 76.75 (telephones per hundred) with urban teledensity at 148.07 and rural at 44.96. Contrast an effective 2 with 76.75,and one gets the vast gap between telephone connectivity and internet connectivity. To obviate the critique that the figure being used here is wireline broadband rather than total connectivity, I draw your attention to two trends evident in Table 2. One, there is a decline in narrowband connections, both wired and wireless. Two, so steep is that decline that the increase in broadband subscribers, both wired and wireless, does not compensate for the decline in narrowband connections, and the total number of internet subscribers actually declines by about 5 million from 259.14 million to 254.4 million.
There are two related points that need to be noted here. The percentage increase of wireless broadband is greater than wired broadband, and TRAI cautions that the speeds on wireless broadband are not good enough. Also, a number of informational and e-governance uses require computers, and those services cannot be effectively accessed on handheld devices. So when we look at these numbers for indicators of digital inclusion instead of just connectivity, the picture is even bleaker. An important caveat, however, is that at least in urban areas there is a likelihood of people accessing the internet for functional purposes not through their own connections but in their workplaces. However, if we consider that only white-collar jobs provide workers with access to the internet in their offices, the numbers show the vast strides that need to be taken for digital inclusion.
Though TRAI does not provide data on internet subscribers by rural and urban areas (unlike data provided for telephone subscribers) we get some indication from looking at the service area-wise numbers (See Table 3 below). The urban skew of subscribers, both narrowband and broadband, can be seen here. So, Mumbai has 14.02 million subscribers while the rest of Maharashtra has 23.21 million. Among the top five service areas are Maharashtra (23.21 million), Tamil Nadu, including Chennai (21.01 million), Andhra Pradesh (18.83 million), Gujarat (17.15 million) and Karnataka (16.56 million). Delhi with 16.42 millionis just a shade below Karnataka, but it has a density above 80. While the disaggregated numbers by city are not provided by TRAI, it bears notice that Chennai, Bengaluru, Hyderabad and the urbanised state of Gujarat clearly indicate the urban skew of internet subscription. This can also be surmised from the service area-wise density data.
Table 3: Service area-wise number of internet subscribers, TRAI, Sept 2014
Total internet subscribers (million)
Total internet subscribers per 100 population
|Tamil Nadu (incl. Chennai)|
Per 100 subscribers are for state and not for service areas in the case of Maharashtra, UP and West Bengal.
Indian government policy
The policy of the union government has been to promote internet use and access. The National Telecom Policy 2012 (NTP-2012) outlines specific targets and parameters. Two specific targets are mentioned as they pertain to broadband: (11)
3. Provide affordable and reliable broadband on demand by the year 2015 and achieve 175 million broadband connections by the year 2017 and 600 million by the year 2020 at minimum 2 Mbps download speed and making available higher speeds of at least 100 Mbps on demand.
5. Provide high speed and high quality broadband access to all village panchayats through a combination of technologies by the year 2014 and progressively to all villages and habitations by 2020.
NTP 2012 further enjoins TRAI to revise the existing broadband download speed of 256 Kbps to 512 Kbps and subsequently to 2 Mbps by 2015 and higher speeds of at least 100 Mbps thereafter.
But as the TRAI Consultation Paper on Broadband notes, the targets have not been met and the ‘country is nowhere near meeting the target for a service which is considered almost a basic necessity in many developed countries’.Since the early years of the Telecom Policy, internet connectivity and later broadband connectivity are part of the overall telecom policy. However, only recently has this become the centerpiece of the Union Government’s agenda with the Digital India Programme. (12)
The policy of the Indian government is necessarily dominated by supply side concerns – provision of infrastructure. On the demand side, it is dominated by one concern – provision of e-governance. While there are three aims of Digital India, providing ‘Digital Infrastructure as a Utility to Every Citizen, Governance & Services on Demand and Digital Empowerment of Citizens’, a closer look at the nine pillars of Digital India that support the three vision areas shows that the demand side is not a key concern for the government. This absence, it needs to be said explicitly, is not necessarily a flaw because governments are firstly meant to provide the supply side deficits, especially where private players do not find it profitable to provide such infrastructure. In fact, the very nomenclature of the vision areas means that the government recognises digital infrastructure as a public utility, and hence its focus on building this network right up to the village.
Since the urban infrastructure is largely in place (though there are those who may argue that optical fibre to the home (FTTH)and optical fibre to the business (FTTB) should be in place to ensure higher speeds, ‘largely in place’ is a comparative statement vis-à-vis the rural infrastructure) the thrust of Digital India has been on ensuring rural connectivity as far as broadband access is concerned. It does provide for changes in rules to facilitate greater urban broadband connectivity.
At this point it may be important to recall that if the internet requires four levels of network – the last mile, the middle mile, the national backbone, and international connectivity – the thrust of Digital India is on enhancing the national backbone to reach rural areas and of providing the middle mile in rural areas. Last mile connectivity is still in the domain of private players or government-owned telephone companies, BSNL,MTNL and RailTel Corporation. (13) Butto ensure that connectivity does reach the village, Common Service Centres (CSC) and post offices as Multiservice Centres (MSC)are being emphasised.
The key targets for the infrastructure are:
- Broadband connectivity to 250,000 gram panchayats by 2016 in three stages of 50,000, 100,000 and 100,000.
- National Information Infrastructure by 2017 by integrating National Optical Fibre Network (NOFN), State-wide Area Network (SWAN) and the National Knowledge Network (NKN).
- Two lakh fifty thousand villages to have CSCs by March 2017
- One lakh fifty thousand post offices to become MSCs by 2016
The other thrust areas for digital inclusion are provision of information and e-governance initiatives. These are supply side initiatives, which have an obvious adoption-level connect. eKranti or electronic delivery of services, falls squarely in this domain and with its focus on education, health, finance, agriculture-related information and judiciary, it can be hoped that there will be a demand for the internet.
Other than these initiatives towards the national backbone and the middle mile, there are some state government-led initiatives in providing last-mile connectivity in public places. While the details are still not available, the decision of the recently-elected Delhi government to provide Wi-Fi or the announcement of the West Bengal government to ensure Wi-Fi across Kolkata are all measures designed to connect the device in the hand of the citizen-consumer to the internet. However, only pricing will make clear what kind of access will be available and for whom.
The most basic areas of concern are access and affordability. The present subscriber base for the internet is extremely low. While the Union government is engaged in providing connectivity, it is still not clear how last mile connectivity will be achieved and at what cost in rural India, assuming that all the targets of Digital India are achieved on time.Recent reports in the press suggest that already the targets have been missed. Only 12% of the target of connecting 25,000 villages by 2014-15 has been met till end of January. (14) Since the onus of last mile connectivity is on private players, what needs to be seen is whether they consider it worth their while to provide this connectivity. Even today the telephone companies provide some kind of wireless access for those with high-end devices paying for higher-end services. But in most rural areas this connectivity is poor.
The anecdotal evidence is that in large swathes of rural India, speeds are wanting or connectivity is non-existent. The reason is not difficult to find. There is a general absence of paying customers in those areas and the cost of either installing or maintaining such a network in those areas cannot be recovered, and so this becomes financially unviable for the mobile companies. In what ways will the economics of providing connections change with the government-provided network is an open question.
The decline in broadband cost over the years provides some hope that with improvements in technology, greater penetration and greater use, the costs are likely to go down. However, a lot depends on the way telephone companies, who still provide the vast majority of connections and are likely to do so in the future, respond to their networks becoming mere carriers. The recent attempt by Airtel to charge differential rates for access to some services on the internet, and hence treat what goes over their wire differentially, is a case in point. The fact that the company had to retreat in the face of immediate consumer opposition does not necessarily mean that such attempts will not be made.
Other than pricing, there is also the possibility that speed over the connections might not be what is being promised. Speed is crucial to realising the full potential of the internet because services and voice and picture componentsare increasing so exponentially that the narrowband or speeds of less than 512kbps are inadequate for a functional use of the internet. In fact, the Indian standard of more than 512kbps for broadband is significantly less than many countries that demand a download speed of above 1 to 2Mbps for the service to be classified as broadband. Even if more functional definitions of broadband are used instead of minimum download speed, the Indian standard is towards the lower end. A recent investigation by the Digital Empowerment Foundation found that in many villages, speeds are half of what has been assured as part of the NOFN.(15) In fact, this report shows that plans towards rural internet connectivity could be fraught with serious implementation challenges.
A related issue is that there is increasing evidence that smartphones will be the device of choice when it comes to accessing the internet. In fact, the plans of telephone companies are predicated on this move that comes on the back of ever-decreasing smartphone costs. The problem, however, is that speeds are still an issue — recall the TRAI consultation document — over wireless networks and most informational and utilitarian uses of the internet still demand either high speeds or large screen devices or both. This is especially true of government and informational websites. To conflate the number of email accounts and social media users with use of internet is to rely on a lowest common denominator base for gauging internet use and digital inclusion.
Since digital inclusion involves both supply side and demand side, or access and adoption, strategies, it is also important to recognise the adoption challenges, which for obvious reasons are not the immediate remit of the government. The basic thrust of government programmes is on providing physical infrastructure. So, all the adoption aspects of digital inclusion still need to be worked on both from those in the government and those in business and not-for-profit enterprises. One possible area to ensure greater adoption is to ensure that schools provide access to the internet, which incidentally is included in the eKranti programme of the government.
However, in the absence of Indian language material and websites, pure access may not be of great use. Even the state government websites provide at best a few layers of material in the state language, while most of the details are only available in English.(16) If one adds the material from non-government institutions, then the task becomes even more difficult. A classic example is of the State Bank of India, which has the largest banking network across the country. It provides an option to read the website in Hindi. But there are no options for any other Indian language, and this from a bank that operates in every nook and cranny of the country.
So, if we revisit the axes of digital inequality – gender, age, disability and language – then digital inclusion in India needs to be combined with the larger societal goals of education, energy, livelihood and health for all.
Aloke Thakore is an independent journalist, researcher, newsroom coach and teacher. He serves as the Hon. Director of the JM Foundation for Excellence in Journalism and has been associated, over the last three years, with a number of research projects on telecom and internet access. He is also the founder-director of Font & Pixel Media Pvt Ltd, a media and education enterprise.
(2) See http://share.cisco.com/internet-of-things.html for an interesting visualization.
(3) See page 11 of World Bank’s Broadband strategies handbook, edited by Tim Kelly and Carlo Rossotto, for a discussion on Broadband as a GPT. For a more general account, please see Bresnahan and Trajtenberg (1992). The working paper of 1992 is available at NBER.
(4) The Telecom Regulatory Authority of India’s Consultation Paper on Broadband (September 2014) calls these local area connectivity, metropolitan or backhaul links, national backbone network and Internet Link and International Connectivity, respectively.
(5) https://i.imgur.com/kXGlWVs.png is a visual representation of internet connectivity put together by Shodan, the search engine for internet of everything. Shodan.io
(6) Not a single Indian language is used by more than .01% of the total websites. http://w3techs.com/technologies/overview/content_language/all Contrast this with the fact that six Indian languages (Hindi/Urdu, Bangla, Marathi, Telugu, Tamil) are among the top twenty spoken languages in the world.
(7 )All the research on digital inequality points to age, income, disability, education and language as barriers. An excellent framework for digital inclusion comes from the Institute of Museum and Library Services, University of Washington Technology & Social Change Group, International City/County Management Association. (2012). Building Digital Communities. Washington, DC: Institute of Museum and Library Services. The framework outlines access principles and adoption principles, which include availability, affordability, inclusive design, public access under access and relevance, digital literacy and consumer safety under adoption principles.
(8) Better life is used not as an index of happiness or satiety, but as a more easily measurable index of time, effort and resources saved in the conduct of the transactions that are mentioned here. Better life because of measurable benefits that accrue from the use of this technology, which reduces barriers of time, place and people.
(10) Please see point 2.3 on page 10 of the TRAI Consultation Paper on Broadband, 2014.
(11) The reason why I have chosen to use broadband as a yardstick of internet connectivity is that narrowband connectivity (and the two terms do not mean the same thing across the world) is showing a downtrend in (June to September, 2014 Performance data released by TRAI in Jan, 2015) and most meaningful use of the internet now demands a ultra 512kbps speed, which in India, per TRAI’s order of July 2014 is the minimum standard to qualify as broadband. Incidentally, TRAI has still not revised broadband speed to 2 Mbps despite the express directive.
(12) The president’saddress to the members of the 16thLok Sabha in its inaugural session clearly articulates this position. “E-governance brings empowerment, equity and efficiency. It has the power to transform peoples’ lives. The backbone of my government’s new ways of working will be a Digital India. IT will be used to drive re-engineering of government processes to improve service delivery and programme implementation. We will strive to provide Wi-Fi zones in critical public areas in the next five years. My government will rollout broadband highway to reach every village and make all schools e-enabled in a phased manner. Technology will be used to prepare our children for a knowledge society. The National e-governance plan will be expanded to cover every government office from the centre to the Panchayat; to provide a wide variety of services to citizens. Emerging technologies like Social Media will be used as a tool for; participative governance, directly engaging the people in policy making and administration.” See http://pib.nic.in/newsite/erelease.aspx?relid=105494
(13) RailTel owns a pan-India optic fibre network on exclusive Right of Way (ROW) along the railway track. It provides broadband connectivity in many parts of the country in addition to modernization of train operations and administration network systems.
(16) As part of their Research on Internet Project, PUKAR (Partners for Urban Knowledge, Action & Research) has been conducting workshops on internet among selected youth in three villages close to Palghar in Maharashtra. Their research into the government websites suggests that whether in the case of information or of government services, it is not easy to utilize the services because of linguistic barriers and also because of user-interface problems. Disclaimer: I am associated with this project as a consultant.
www.digitalequality.in, February 2015