Tag Archives: Mahesh Uppal

13 Jan

Free Basics: Regulatory principles versus Ideology

Free Basics

As published on Firstpost.com

Should Facebook’s Free Basics initiative be stopped immediately? A response to this question cannot depend on claims, either of its promoter – Facebook – or its critics. And, the issue is not whether Facebook’s goal is to expand internet access to the poor or to further its business. It is whether there is any evidence that Free Basics violates any existing rule, harms consumers or other players in the market. An economic regulator, such as Telecom Regulatory Authority of India (TRAI) will need to settle this, using data and well established tests, not petitions that support or oppose Free Basics.

Read More

11 Mar

Spectrum auction: Modi government’s approach is out of sync with Digital India plan

*As featured on Economic Times on March 8, 2015

- By Mahesh Uppal

By Friday evening, bids in the ongoing spectrum auction had reached Rs 77,000 crore, or over $12 billion. They could rise substantially before the auction concludes. The bids are great news for the central exchequer. Regrettably, the bids also suggest the government’s approach to spectrum is out of sync with its ambitious plan for a Digital India. More on that later.

Despite the spectacular bids, the government would be nervous about the decision of the Supreme Court which has stayed the publication of auction results and the award of spectrum to winners. GSM operators have challenged some of the provisions of the auction. We will know more on the next hearing on March 26.

The spectrum being auctioned includes frequencies in the 800, 900, 1,800 and 2,100 MHz band. Except for the 2,100 MHz band, almost all of the spectrum being auctioned for any service area, or circle, is in use by operators whose 20-year licences will expire in the next few months.

No Surprises There

The high bids are not surprising. Several companies face unenviable choices as they have large investments and revenues to protect. For instance, Reliance Communications could have to discontinue services in up to seven circles, if it can’t win back its spectrum. The auction has bearing on up to 75% of Idea’s revenues. Vodafone faces serious pressure in six circles. Reliance Jio, which has a nationwide 4G spectrum, has deposited the highest earnest money. It can wrest spectrum from current incumbents as well as raise their costs of acquiring spectrum.

It is easy to understand the rising bids for 900 MHz spectrum. It is by far the most valuable because lower frequency signals can cover longer distances and therefore require fewer telecom towers. The spectrum also supports 3G data services. It was allocated first when mobile services were launched 20 years ago. Early entrants would be especially keen to retain high-income subscribers acquired when call rates were several times higher than today. This spectrum thus allows operators to combine their immediate commercial goals with their future ambitions in mobile broadband.

The auction of 2,100 MHz (3G) spectrum saw less action in the beginning of the auction. But this is changing and bids can be expected to rise further. This spectrum is globally harmonised and companies have repeatedly sought more 2,100 MHz spectrum. They had offered to accept partial assignments in areas where defence needs some of it and to bid for spectrum before it is fully vacated. So, one can expect more interest in 2,100 MHz once bidders take a final call on the price they are willing to pay for the 900 MHz spectrum.

The demand for wireless spectrum reflects the fact that 95% of Indian users rely on mobiles for telephone calls and internet. Further, no other country has more operators in any service area (called circles in India). The limited spectrum must cater to more users as well as operators than elsewhere.

Thanks to the high competition, bids may not cause a dramatic rise in rates, though the prices are climbing steadily in recent months. However, the operators can be expected to prioritise high paying customers over growing markets — especially weak ones e.g. rural or broadband services where much more investment is needed. This is no less worrying.

What about Digital India?

The high sums do not mean spectrum auctions are a bad idea. On the contrary, auctions are the only transparent way of dealing with a demand for spectrum that far exceeds its natural supply. However, better auction designs are possible and eminently desirable. For instance, regulators often mandate that winners deploy broadband services in a prescribed short time or face penalty. This could have resulted in lower bids but better use of spectrum without compromising on transparency.

India’s overwhelming reliance on wireless is all the more reason to manage spectrum strategically. Fortunately, unlike other natural resources like coal or water, it is totally inexhaustible. Its use today has no bearing on the interests of coming generations. This allows much greater flexibility in managing it.

It is self-defeating to design auctions solely to maximise upfront revenue. Unfortunately, the government has chosen poorly. It went against the advice of a statutory regulatory body on critical aspects of the auction. It failed to accept the recommendation to auction more 2,100 MHz spectrum despite declaring that it had an agreement that defence would release more soon. It couldn’t deliver on the spectrum sharing and trading guidelines due in December last. That would have eased the pressure on bidders. The government seemed to have prioritised its revenues over a coherent regime for spectrum.

A government that believes a “Digital India” can be an engine of growth and governance cannot justify artificial barriers to efficient use of scarce spectrum. The success of Digital India is critically dependent on how well India manages the allocation and price of spectrum.

08 Jan

What to do about Net Neutrality?

By Mahesh Uppal

The Mobile operator Airtel’s recent decision – hastily withdrawn – to charge more for internet based telephony services goes against the principle of Network Neutrality.

The broad principle of Network Neutrality makes sense. It attempts to impose on internet service providers (ISPs) an obligation similar to the one on other utilities like power, roads, railways, etc.  It requires that a service provider not discriminate between users or the type of usage. So, just like it is unacceptable if a player who runs railways or roads, determines if, when or where you travel, so should an ISP not have a say in your use of the internet.

India, like most countries, does not mandate Network Neutrality explicitly.  However the licences of telecom operators like Airtel, Vodafone, Bharat Sanchar Nigam Limited (BSNL) etc. who provide the bulk of access to the internet require that the holder “shall not in any manner discriminate between subscribers and provide service on the same commercial principle”.

The Network Neutrality debate acquired new energy last November, when US President Obama reiterated his strong support for the principle. He said: “We cannot allow Internet service providers (ISPs) to restrict the best access or to pick winners and losers in the online marketplace for services and ideas.” He opposed attempts at blocking websites selectively, throttling or slowing down some traffic and paid prioritization of traffic. He presumably had in mind companies like Comcast, AT&T and Verizon who own and dominate the underground cable and optical fibre lines that most Americans use to access the internet. The latter have argued that the investment needed to accommodate increasing traffic, especially video, was becoming unsustainable. They want the right to be able to prioritize some types of traffic. Obama, like other supporters of Network Neutrality, sees such prioritization as potential abuse of dominance. He called on Federal Communications Commission (FCC), which regulates telecommunications and broadcasting in the US, to devise Network Neutrality rules.

Are Network Neutrality rules relevant for countries like India? Is abuse of market power by network operators rampant, serious or imminent? Is Network Neutrality relevant in cases where the reach and capacity of networks is limited?  Could the rules be an overkill? Should Network Neutrality be recommended or obligatory? Should we treat it akin to a constitutional right, for example to life, property or free speech?

In my view a doctrinaire or rigid approach to Network Neutrality is illogical and risky. This is especially true for developing countries like India whose data or Internet markets differ qualitatively from those of the US. In India, no company providing internet access has anything close to a monopoly in the market.  Fixed line market is admittedly dominated by BSNL. However, the latter has far fewer subscribers than wireless operators – on whom a majority of users rely for access to the internet.

There is another important difference between the US and India. The predominant mode for access to internet in US is fixed line and in India, wireless.  It is far easier to expand capacity of fixed networks in US than the wireless ones in India. Increasing the capacity of cable and fibre networks depends mainly on availability of funds. However, for wireless networks funds alone are insufficient. It depends on availability of needed spectrum and the willingness of government to release it. In the case of India, operators cannot buy or lease it from current users of spectrum.

Network Neutrality therefore implies different burdens for fixed and wireless players as well for players with and without market power.  So while a government or regulator could require a dominant fixed line player to expand capacity to accommodate increasing levels of traffic it could not realistically require capacity-strapped wireless operators to do so.

In India an added issue in mandating Network Neutrality is the huge regulatory costs – licence fees, spectrum usage charges – charged from network operators amounting to roughly 30% of their revenues.  VOIP and other Over-the-Top (OTT) services pay no such fees despite the fact that phone services like Skype, Viber etc. or messaging services like WhatsApp are functionally similar – in some ways, even more powerful than the corresponding services of mobile operators. This ‘unfair competition’ poses a different type of argument against network neutrality in India.

A key dichotomy in implementing Network Neutrality is that intended beneficiaries could potentially be victims too. While it seeks to assure smaller players of equal access to networks, it provides no guarantee that bigger players, especially social media with its much wider appeal and sophisticated technologies, will not grab bulk of the bandwidth, as many believe, they do currently.  This is especially true for wireless networks.

It is notable that bigger players like Facebook, Google etc., which are strong advocates of Network Neutrality are making exceptions themselves. In many instances, in India and in other developing countries, they have made deals with mobile operators that favour their services over others’. In some cases, for example, access to the services of one player, say Facebook, is free, but not for websites or services of players, without comparable commercial clout to enter into similar arrangements.

In my view, the central issue in public networks is to fair access and not the sanctity of the Network Neutrality principle, per se. Sometimes Network Neutrality might hurt more than it can help. This is an issue in almost all types of networks.  For instance, in case of road networks, big trucks are treated differently during peak traffic hours. There are toll roads for users who want to save time. In India, the “Rajdhani” and other trains exclude passengers from most stations on their route. In electricity, we have differential pricing for low and heavy users. Can we argue that these breaches of Network Neutrality are entirely unreasonable?

However, it would be patently unfair if all roads were toll roads or all trains Rajdhani’s. We cannot divorce Network Neutrality from the larger context of public policy.  ‘Violations’ from Network Neutrality must be the exception and not the rule.

Network Neutrality means ensuring that those who own or control networks behave reasonably. Such players frequently have incentives to favour certain type of users or usage that provide higher returns. Since network businesses have little or no competition, these players have the potential to abuse markets through arbitrary pricing. This can hurt users and sometimes thwart new entry to the markets.

It is these cases that economists call “market failure” where Network Neutrality is most important. It poses much the same questions that regulators face when deciding whether or not intervene in tariff setting or whether licensed operators must be mandated to serve rural areas, even if it is unprofitable to do so.

Network utilities have profound implications for public interest. Obama’s focus on ensuring no one is disadvantaged in their use of broadband networks in the US is no less relevant in India.  However, there is a big difference between the reach of broadband networks in two countries. The priority for India is to deploy the networks while for US it is to improve access to the broadband networks already deployed.  It would be counterproductive if poorly implemented Network Neutrality gave the operators an excuse to make it difficult or expensive to use popular services such as VOIP telephony. India must ensure that no sets of players carry unfair regulatory burdens for providing competing services and users of popular services have unbridled access to technologies that provide powerful and cheaper services.

Airtel’s approach to VOIP tariffs highlights the need for the government and the Telecom Regulatory Authority of India (TRAI) to collaborate on a model of Network Neutrality that is relevant for India.

*Views expressed here are personal

As featured on Thinking Aloud supported by IAMAI

12 Nov

Net Neutrality

I believe, Obama is broadly- but not entirely- right about net neutrality. ISPs – or telecom operators who are the bigger ISPs in most cases- should not in general be able to discriminate between different types of content or its creators.

However, the problem is when net neutrality is applied to wireless networks where capacity is a bigger issue because of spectrum scarcity. Wireless networks simply do not have capacities comparable to cable/fibre networks. India is a classic case where operators have a fraction of the spectrum that their peers in other parts of the world.

Developing countries need last investments in infrastructure. Net neutrality may be counterproductive if this became the excuse for operators not to expand networks to the large regions which are still uncovered.

In any case, there are several cases where content players, like Facebook, Google etc,- understandably supporters of net neutrality- are working with mobile operators to protect their commercial interests. They have arrived at arrangements, across the world, that would not be strictly in line with network neutrality.

The problem with net neutrality is at the edges. We all want free movement on roads but yet, can accept a limited number of toll roads. Similarly, we must be willing to accept that a fundamentalist position on net neutrality might hurt those whose rights we want to protect.

This is a challenge for regulators. They will need to balance consumer interest in content with their huge stake in growth and capacity of networks.

- By Mahesh Uppal

28 Jul

Economic Times – Department of Telecommunications may use Telecom Regulatory Authority of India’s data to reach uncovered areas

“Accurate and robust data will be invaluable to estimate India’s rural mobile coverage. It can also be useful for designing future auctions as it will help specify realistic rollout obligations for telcos that win spectrum through those auctions”

Click on the PDF icon to read more:


Quotes from Dr. Mahesh Uppal

23 Apr

Fixed Fee is the Fix

On April 17, the Supreme Court upheld the government’s plea that Comptroller and Auditor General of India (CAG) was within its rights to audit the accounts of private telecom companies.

The problem with the judgment is not that Supreme Court is mistaken, but that the ruling deals with the symptom of a more serious problem viz. computing licence fees as a proportion of operator revenues. There is an obvious incentive to understate revenues and escape payments to the government.

This concerns the CAG. Telecom licence fees are a large contribution to the exchequer. And, as the Supreme Court has confirmed, CAG’s s mandate includes government receipts as well as expenditure.

However, CAG’s new task is huge and difficult. The telecom sector has revenues of roughly 22,000 crores (US$38billion) spread around 13 operators, in 23 circles involved. Further, spectrum charges are linked to the type and amount of spectrum held by the operator as well as the revenues generated.

CAG’s staff will face the same challenges that India’s tax inspectors do with income tax returns. Good ones will struggle to finish the tasks in time. Bad ones will help companies to escape payments. This is hardly an improvement; it’s more like business as usual.

It is not the Supreme Court or the CAG, but the Department of Telecom (DOT) that can solve the problem.

The way out is to delink fees from operator revenues. The answer is to set fixed fees. The amount could be a percentage of total sector revenues. DoT can set the percentage to enable government to protect its revenues.

How much would individual companies pay? A simple, logical and workable way to determine a company’s fee would be to divide the total amount of fees in proportion to the amount of spectrum held by it. So, a company which holds a quarter of spectrum allocated, would contribute quarter of the fees.

What if a company does not own spectrum? It would make sense to charge a prefixed nominal fee or to forgo it altogether, since all players without spectrum, even today pay a less than 5% part of the sector’s total fees.

A fixed fee will protect government’s interests without increasing the incentive for fudging. Indeed, it will encourage greater efficiency. Efficient companies will retain more of their revenues, inefficient users of spectrum as well as hoarders will pay more. That is good for the sector and for the overworked auditors.

Watch this space for details.

By Mahesh Uppal

Copyright © 2016 Com First All Rights Reserved