9 January 2016

End-to-end ka The End: Our response to TRAI Consultation Paper on Differential Pricing for Data Services

A report by

Abstract

The Telecom Regulatory Authority of India (TRAI) invited comments on their consultation paper No. 8/2015 on differential pricing of data services. The Internet Democracy Project submitted comments to TRAI, outlining why differential pricing of data services in the Indian mobile telephony market further jeopardises the goal of wider access to a free, open and secure internet, and runs against consumer welfare and competition.

Download Report

At the outset, TRAI notes, in the consultation paper, that its regulatory approach has been one of forbearance, and regulatory interventions whether ex-post or ex-ante, are exceptions which are made where consumer interest has to be protected and orderly growth of the sector ensured.

TRAI also identifies that its mandate is to strike a balance between achieving wider access to the internet on the one hand and ensuring that the manner in which this wider access is provided does not violate the principles of non-discrimination and transparency on the other [paragraph 3 of Consultation Paper].

For providing wider access, differential pricing of data packs has been touted by some as a workable business model. While it would succeed in widening the market for data services by enabling more people to access a limited set of websites and services, and this at reduced or no cost, this would create an inferior quality of access. In this submission, we argue that further empowering Telecommunication Service Providers [TSPs], or enabling platform entities to curate a menu of websites and services at an attractive price does not bode well for either the principle of non-discrimination or for transparency.

Question 1: Should the TSPs be allowed to have differential pricing for data usage for accessing different websites, applications or platforms?

A: In oligopolistic markets of limited resource of spectrum in India, TSPs have significant market power and they should not be allowed to have differential pricing for data services for accessing different websites, applications or platforms.

A. Price discrimination violates the principle of non-discrimination

The Telecommunication Tariff (Thirty Third Amendment) Order of 20041 clarifies the interpretation of the principle of non-discrimination. It says that differential tariffs assuming the nature of vertical price squeeze would not be permissible.

“An operator with significant market power can often squeeze the margins of competitors by raising wholesale prices (upstream market) paid by competitors while at the same time lowering retail prices (downstream markets) on competitive services. Such ‘squeezes’ on the margins of competitors imposed by the pricing strategy of the operator with significant market power could materially affect competition.”

  • Competition among applications/content providers is materially affected

In the Indian mobile telephony market, TSPs are oligopolies with significant market power, and operating with finite competition, providing access to a limited resource. Price discrimination vests enormous decision-making power in these private players, whose choice of the suite of applications and websites will not necessarily be aligned with consumer interests. Making them arbiters of usefulness and relevance to a large number of persons with potentially diverse needs would not necessarily align with public interest, and unless there is perfect competition, regulatory interventions will be needed.

TSPs may choose to tie-up with affiliated companies in the application/content provider layer, over other competing players who may not be similarly advantaged. TSPs could require that content providers be exclusively available over their price differentiated packs and not over data packs of other networks. Besides, a large number of non-commercial endeavors like personal websites and campaigns are not likely to participate in such platforms, and are shut out from accessing a large number of customers/users.

The same harms that vertical integration brings can materialise in scenarios where TSPs are tying up with content providers that TSPs are not previously affiliated with. TSPs could potentially use price discrimination to extract “oligopoly rents” from the content providers’ side of the market, which is a highly competitive side. This might not necessarily be a monetary price that is extracted, but might for example entail requiring applications and content providers to not offer services like VoIP that threaten TSPs’ traditional voice offering. TSPs prioritising content for their differentially priced cheaper data service packs gives TSPs an incentive to throttle non-participating content. TSPs in other parts of the world have been disingenuous in this regard.2 By increasing input costs for content providers and requiring them to apply to participate in platforms and conform to technical specifications that might not be directly related to consumer interests, TSPs could impose non-monetary burdens.

The consultation paper points out that there may be more than one form that differential pricing might take. The different forms of data tariff offerings that have found mention3 in the paper are all united in their anti-competitive effect for reasons explained above.

  • Mediated access creates information control and this discriminates between the quality of access that those who can pay and those who can’t pay get, threatening consumer interest.

The National Telecom Policy of 2012 aims to ensure that India transforms the socio-economic scenario through accelerated equitable and inclusive economic growth by laying special emphasis on providing affordable and quality telecommunication services in rural and remote areas.4 The National Telecom Policy, thus, does not see any conflict or compromise between affordability and quality of telecommunication services.

If the Internet has proved to be empowering for many of its users, its free, open and secure nature is at the core of this. In India, the majority of Internet users will be accessing the Internet through a mobile phone. Though the Internet experience of consumers is increasingly mediated by corporations and governments irrespective of the mode of access, this is especially true for mobile Internet. This means that if the empowering potential of the Internet is to reach mobile-first and mobile-only users as well, ensuring that their experience of the medium is as free, open and secure as possible becomes essential.

It could be argued that app stores which are tied to platforms like Android or iOS limit competition, and price differentiated or zero-rated platforms will be no different in effect. The mobile app environment is surely not as free as the internet experience available through browsers on personal computers. But this difference in the free, open and secure nature of the internet between what is possible over mobile phones and on personal computers via browsers will be further exacerbated. At the moment, users atleast have a choice amongst a range of applications, notwithstanding the barriers app stores and ‘rating’ of apps pose.

  • Consumer behaviour is shaped by first-movers and switching costs for consumers will be high

Any regulation aimed to benefit a billion people should improve awareness of the benefits of a free, open and secure Internet. For users of data packs, this means the awareness of ‘the ability to connect, consume, create, collaborate and correct (content)— all in equal measure’5, and not be limited to passively consuming certain popular services.

Introducing new users to a limited set of websites/applications makes them likely to be reluctant to switch to a competitor. This is the nature of a lot of applications and services on the internet. This is to a certain extent even true of mobile phone numbers, where there are costs in switching to other network providers. So even in markets with sufficient competition, content providers who are first movers would have an advantage over the late-comers and winners could be decided dependent on who the dominant TSPs tie up with.

  • Making TSPs or platform entities gatekeepers would have serious implications for freedom of speech

In many parts of the world including in India, mobile internet has played a big role in active citizenship and has facilitated people’s political participation. If TSPs or platform entities are gatekeepers to a large number of persons of similar demographic, there is a real possibility of blocking certain kinds of news media or content when it suits the TSP/platform entity to do so. The existence of such points of control is itself a matter of concern. This has precedent in different parts of the world. During a labour dispute in 2005, members of the Telecommunications Union in Canada were unable to access a website disseminating the union’s views because the ISP Telus had blocked access to it.6 Such situations could be worse with vague and opaque participation guidelines.

Classifying differentially priced data packs on the basis of pre-decided criteria is better than deal-based differential pricing, but would still not be desirable as that kind of information control would make the data services available only marginally better than pre-existing cable TV networks. This also runs the risk of treating consumers differently from service or content providers by assuming that consumers cannot also provide content or services, which may actually sanction network operators to reduce the choices available to consumers as well as undermining the empowering potential of the Internet for its users.

B. Price discrimination affects innovation

Regulation should seek to incentivise, and not disincentivise innovation towards lower-bandwidth applications. By TSPs precluding service types like VoIP, development of lighter VoIP applications are discouraged as they would still threaten the TSP’s revenues. Further, incentives for start-ups and existing businesses to build services specifically targeted at those with limited bandwidth would also be affected, as they could instead acquire this consumer base by tying up with TSPs to be on differentially priced cheaper platforms.

Having a level playing field and non-preferential treatment also encourages new users to contribute content apart from being passively consumers of the internet, leading to innovation that understands local needs. A seamless experience that the Internet currently provides, encourages exploration and discovery on the part of users as well as furthers innovation on the part of business, the social sector and government.

Further, if the target population for zero-rated packs are indeed low-income groups including illiterate parts of the population, local language material consumable through the medium of videos should not be discouraged, but worked around.

While it is reasonable to expect that new users of differentially priced data packs would find value in paying for access to the wider open internet and eventually pay and get out of the price differentiated packs and walled gardens, there is a likelihood that a large portion of the Indian population would not be able to afford the larger internet and access it. TSPs would likely increase prices of all-access packs if ‘altruistic’ and ‘access-boosting’ motivations are all located in cheaper or zero-rated packs.

C. Price discrimination violates the principle of transparency

As price differentiation is done in the name of improving affordability for semi-literate and illiterate populations, it is important to ensure that such packs don’t result in fraudulent or opaque billing of users. While TRAI can issue strict guidelines imposing penalties to TSPs/zero-rated platforms, its capacity to enforce these penalties might be diminished from the heavy regulatory burden this brings. This has previously been the case with measures taken with respect to VAS guidelines.7 However, transparency is important not only in the context of billing, but also important when it comes to terms of use and data protection. In a model where users are not charged money for the data they consume, platforms curating the menu of websites and services can be expected rely on harvesting user data from consumer acquisition, and this often happens under extremely non-consensual and opaque ways.

In conclusion to the question, in the interest of consumer protection and competitiveness and innovation in the internet ecosystem, differential pricing should not be allowed. There is a lot of room for mis-leading policies and disingenuous, surreptitious prioritisation of content and applications. Neither TSPs nor platform entities should be allowed to indulge in price discrimination.

Question 2: If differential pricing for data usage is permitted, what measures should be adopted to ensure that the principles of non-discrimination, transparency, affordable internet access, competition and market entry and innovation are addressed?

A: As explained in the response to Q. 1, the very nature of vesting carriers with power to determine what the suite of applications would contain is discriminatory and the sustainability of such endeavours is premised on lack of transparent terms of access, as the harvesting of user data and behaviour is what the applications would be after.

While certain services would be affordable to access, the benefits of a free and open internet would be lost. Competition and market entry would be compromised, and input costs for new application providers would be higher. Besides, only commercially oriented services would reach the users of such services. Innovation, especially in the area of creating lighter and affordable technologies for access would be disincentivised.

The principles that TRAI aspires to uphold are all at the risk of being compromised.

Question 3. Are there alternative methods/technologies/business models, other than differentiated tariff plans, available to achieve the objective of providing free internet access to the consumers? If yes, please suggest/describe these methods/technologies/business models. Also, describe the potential benefits and disadvantages associated with such methods/technologies/business models?

A: There are alternative methods, technologies and business models through which the problem of affordable access can be tackled, without compromising on the free, open and secure nature of the internet.

Earned data/equal rating As explained by the Alliance for Affordable Internet8, earned data is where “instead of directly purchasing data, the user receives data in exchange for performing some action. Such actions include , completing surveys, or other marketing services on certain apps. It can also include purchasing specific services or handsets from carriers. Typically, this data can be used to access any site or service, though in some cases the data is only to be a used for a specific site.”

Grameenphone, in partnership with Mozilla provides 20MB of free data each day for customers who purchase a Symphony handset and use the Grameenphone app Wowbox.9 This is an example of a business model that can be sustainable in lowering data costs for users.

Gigato is an app offering in India that offers data rebates to upon using particular apps for a certain amount of time.10 Apps like mCent let users earn data in exchange for completing surveys, watching videos or contributing to other marketing-related activities.11

Content providers who can afford to spend such promotional costs have an advantage over their competitors who cannot afford this input cost. However, in giving a rebate of free data to browse as the user pleases, the harm of walled gardens is eliminated. Unlike classic price differentiation, equal rating would not work to obfuscate awareness of the internet as a medium also for decentralised alternatives to platforms and corporations.

Spectrum use Efficient use of spectrum by regular audit of spectrum usage and de-licensing additional frequency bands for use in backhaul would help in the goal of providing wider access to the internet. These reforms have also been on the cards for a while as recorded in the National Telecom Policy 2012.

The use of white spaces should be promoted without causing harmful interference to the licensed applications in specific frequency bands.

The ultimate profitability of TSPs should not be a value against which all other values are measured. At a time when numbers indicate that profits of telcos are increasing and 100 million users have come online in the last year12, these factors should be considered before accepting claims by telecom companies that they need more incentives to build better infrastructure.

Question-4: Is there any other issue that should be considered in the present consultation on differential pricing for data services?

A: Although it might be beyond TRAI’s mandate, options like utilising the Universal Service Obligation Fund for providing capped data packs in the form of direct benefits might be more suited to solve the problem of wider access to the internet and deserve serious consideration by appropriate authorities.


  1. See http://www.trai.gov.in/WriteReadData/UserFiles/mpci/amend8dec.pdf 

  2. See https://www.techdirt.com/blog/netneutrality/articles/20151231/18201233216/t-mobile-is-fat-outlying-throttling-video-even-though-it-says-not.shtml 

  3. Paragraph 8 of Consultation Paper “Under one form, the service provider selects the content, which is offered free or bundled together at reduced rates. In another form, one content provider creates a platform where other content providers can apply, and be selected. The platform creator then partners with service provider(s) to provide free internet access to participating content providers, for the subscribers of those service providers.” 

  4. See http://www.trai.gov.in/WriteReadData/userfiles/file/NTP%202012.pdf 

  5. See https://medium.com/hacks-hackers-africa/taking-free-basics-in-kenya-for-a-spin-87d2a6e9e5a0 

  6. See www.publicknowledge.org/pdf/pk-net-neutrality-attach-20060206.pdf 

  7. See http://www.medianama.com/2015/03/223-trai-regulations-on-vas-need-more-teeth-amba-kak/ 

  8. See http://a4ai.org/the-impacts-of-emerging-mobile-data-services-in-developing-countries/ 

  9. Ibid. 

  10. Ibid. 

  11. Ibid. 

  12. See https://twitter.com/rsprasad/status/682130503688794112 


Related Issues