Flagship Research Quarterly of the Indian School of Business (ISB). Find out more

Sign up  |  Log in

Past Issue • Jul-Sep 2013

Media Plurality And Net Neutrality

In this paper, Professor Rohit Prasad contrasts the regulatory approaches in traditional  media and the Internet and argues that, a new approach is required in the increasingly converging media world. This regulatory approach must synthesise diversity of media content, ie, media plurality and the free play of consumer choice and service innovation on the internet, ie, net neutrality and promote community-generated content.

Traditional media and the Internet are both content businesses, although  the  Internet   is additionally an applications  business. In this paper we deal with the regulation of news  generated  on television and on the Internet. We assume the aim of media regulation is plurality,  i.e., media should reflect a diversity of opinion, generated by a broad  spectrum  of citizens and  shared widely with relevant  sectors  of the population.  We start by delineating the value chain of a generic content  business, present the variations in the case of traditional  media and the Internet, identify the different regulatory approaches used in each and suggest a  way forward based on  a convergence  of traditional and new media.

 

article 6_Figure 1

 

Value Chain of a Content Business

Every content  business consists of a content creation layer, a content transmission layer (intermediary) and an end user. This generic value chain acquires different  hues in the context of the different  forms of TV broadcasting: terrestrial,  cable, direct-to-home (DTH),  broadband and wireless. In terrestrial  broadcasting and DTH, signals are sent directly  to the user’s television and are unscrambled using  a set top box. In cable TV, signals are sent  to a multi-service operator  (MSO) who unscrambles them and sends them to local cable operators (LCOs),  who send them forward to consumers. In broadband, telecom operators install optic  fiber cables to users’ homes and offer a “triple- play” service of voice, data and TV connectivity.  In wireless, telecom operators provide high-speed Internet connectivity, which can be used to wirelessly access TV programming  being  streamed on the Internet.  The value chain in the case of cable TV, still the most popular mode of watching TV, is shown in Figure 2:

 

article 6_Figure 2

In the  case of the Internet, the content created by the content provider is hosted by a hosting  service provider (HSP) who has an agreement with an end user connectivity provider (ECP) to reach the end user.  The  end user accesses the content on a device such as a mobile phone or a tablet. The value chain in the case of the Internet is shown in Figure 3 (the device vendor is left out for ease of exposition):

 

article 6_Figure 3

 

Two very important and interrelated factors determining  the nature of the value chain are the manner in which different entities in the value chain are compensated and their role  in deciding  the content that the user consumes.

 

article 6_Figure 4

 

In the cable TV value chain, prior to digitisation, the content provider paid the MSO, the end user paid the LCO and the LCO paid the MSO. The advertisers paid the content provider. The LCO created various packages of programmes,  based on what they could afford to buy from the MSO, and sold them to the end user. The MSO and LCO had the freedom not to carry certain programmes on their network in order to conserve scarce capacity, or, conversely, to charge a  premium for pushing certain content  providers on basic bands. After digitisation, it has become easier for the MSO to track the programmes the end user watches. Hence, the LCO cannot underreport revenues. Part of the revenues from  the end user can also flow to the broadcaster.

On the Internet,  the content provider pays the HSP, and the end user pays the ECP. The advertiser pays the content  provider.  The  HSP and the ECP usually have  a “bill and keep” arrangement, under which  they provide free services to each other and allow  each other to charge content providers and end  users respectively.  A controversial issue relates to whether ECPs can charge content  providers and on the measures they can take to conserve scarce capacity on their networks.

 

article 6_Figure 5

 

Regulating the Value Chain in Media and the Internet

The description  of the value chain  illustrates key aspects of the  regulatory approach in traditional media and the Internet.  In traditional media, the aim  is to curtail the market power of the content providers, in order to ensure that a wide range of opinion  is available to the viewer. The market power of the intermediary is not the focus of the regulatory scanner.

The  broadcaster  must  stand  ready  to provide its content to any intermediary. It is not mandatory, in many cases, for the  intermediary to carry the content.   For instance,  the Indian Broadcasting and Cable Services Interconnection Regulation 2004 has a  must-provide clause for broadcasters but no must-carry clause for intermediaries (ASCI 2009). With  digitisation, there is perceived to be no capacity crunch for the intermediary, and hence, the intermediary cannot  refuse to carry the content  of any broadcaster. However, it is considered acceptable for the intermediary to create a portfolio of plans from which viewers choose. The accretion of market power  resulting  from vertical  integration  between the content creator and the distributor is controlled through limits  on cross-holding – for instance,  in India,  a broadcaster cannot own more than 20% of an MSO (TRAI 2009).

Regulation of the Internet,  while a relatively new phenomenon compared to traditional  media, takes a different approach to regulating the content provider and the intermediary. One of the influential  schools of thought in this area is net neutrality.  The aim of net neutrality  is to let the content and applications that emerge on the Internet  be decided entirely by the interplay of consumer choice and service innovation. It seeks to circumscribe the role of the intermediaries between the content provider and the end user, especially that of the ECP. In the  case of fixed line access, the ECP may be a natural monopoly, and in the case of wireless  access, an oligopoly  with  a measure of market power.  This movement  is based on the “end- to-end”  design principle  (Saltzer et al 1984). As per this principle, the control and intelligence functions must reside largely with  users at the “edges” of the network (content provider and end user), rather than in the core of the network itself (intermediary). It is believed that the regime of net neutrality  has driven the amazing innovation  witnessed on the Internet.

Net neutrality proponents are mindful  of the fact that vertical integration between an intermediary and a content  provider  can also militate  against the end- to-end design principle.  Hence they look askance at discriminatory  behaviour  arising  in the  context  of vertically integrated entities. The ECP cannot refuse to carry any content or application based on  the amount  of bandwidth  required  or other  technical characteristics of the transmission  (such as required mouth to ear delay). Further,  it cannot charge content providers  and, in some extreme  versions of net neutrality, cannot use traffic management techniques to manage network  flow.  Investments required  to carry content  are to be recovered from  end  users based  on  non-discriminatory criteria like volume of download. Content providers are required to pay their HSP who has a bill and keep arrangement with the ECP.

 Regulation of the Internet, while a relatively new phenomenon compared to traditional media, takes a different approach to regulating the content provider and the intermediary. One of the influential schools of thought in this area is net neutrality. The aim of net neutrality is to let the content and applications that emerge on the Internet be decided entirely by the interplay of consumer choice and service innovation.

The  rules on  net  neutrality  announced in the United   States (US) in 2011  require  transparency in network  management by intermediaries  and prohibit blocking of content or applications by fixed line broadband service providers (who  do not have capacity constraints) and discrimination  by mobile service providers against applications that  compete with  their  own offerings, e.g. Skype. Netherlands is the first country to apply net neutrality in full for both landline and mobile operators.

Net Neutrality versus Media Plurality

The discourses on net neutrality  and media plurality are interlinked  because both industries are content- based and therefore,  share common  principles.  In economies like India, the growth of the Internet  is at  a nascent  stage but is gaining  in importance.   As the Internet  becomes a channel for the distribution of all kinds of content, including  news, one cannot have media plurality  unless there is a reasonable environment  of innovation  and  openness on  the Internet. Similarly, injunctions on net neutrality would  be diluted  unless the traditional  channels of media also deliver pluralistic viewpoints.

Given the  absence of any discussion  related  to the content provider or significant  elements of the value chain  such as device manufacturers, the implicit assumption  of the  net  neutrality  approach is that market  power  resides in the intermediary  but not in the content provider or the device manufacturer. Alternatively, the premise may be that such market power may exist, but barriers to entry are low  and that inefficiency or lack of innovation will be punished by new entrants. Inherent  to this reasoning is the hypothesis that the end user, unimpeded  by a market power exercising intermediary, is qualified  to assess the value of different content and application offerings. In other words, there is no asymmetric information between the user and the content provider. The assumption in the media plurality discourse on the other hand is that market power lies with the content provider and not the intermediary. A further assumption is that either  consumers are not always capable of making out “good” from “bad” content or, even if they can that the barriers to entry for new broadcasters would prevent media plurality.

The differing  approaches of media plurality  and net neutrality can partly be explained by the different histories of the traditional  media industry and the Internet. The history of the regulation of traditional media goes back to the early years of the 20th  century, with roots in print media and terrestrial TV. The role of the intermediary  in the deliver y  of print news and terrestrial  TV was negligible. Perhaps due to this reason, not  much attention  was paid to monitoring the intermediary, even after the intermediary became a vital cog in the value chain with  the coming of cable TV, broadband and DTH.

On the other  hand, at the time  of the advent of the Internet,  the most entrenched entities in the landscape were  the  intermediaries  who  provided the dial-up connection – the telephone companies. Telephone companies had traditionally been under the regulator y scanner and thus, almost by default, drew attention in the Internet industry as well.  Preventing them from taking  advantage of their   position   as a bottleneck  monopoly,  i.e., the sole conduit to the end user, was seen as the primary aim of regulation.

Convergence of Regulatory Approaches

The  stances in both  the traditional  media and the Internet  need to be broadened. As regards the assumptions  of net neutrality, certain content/ application  providers, for instance  Google,  have significant market power that the end user may not always be able to sense. For  instance,  search returns may  be taken  as having objective truth  value, while in fact,  they may only  reflect  the  wisdom  of one particular algorithm for choosing relevant websites.

There are high barriers to entry on the Internet, even without marauding intermediaries. Setting up an Internet Protocol television (IPTV) channel requires significant investment. Further, entities other than content providers and intermediaries,  such as device manufacturers,  have significant  market  power. Apple  adopts a “walled  garden”  approach whereby it provides  a seamless end-to-end  user experience including  content, apps and connectivity. Thus, it becomes an important  filter  for the kind of content its loyal customers consume.

 

 In economies such as India, the growth of the Internet is at a nascent stage but is gaining in importance. As the Internet becomes a channel for the distribution of all kinds of content, including news, one cannot have media plurality unless there is a reasonable environment of innovation and openness on the Internet.

 

The  assumptions of media plurality  are also questionable. Some intermediaries may have significant market power that could rival that of the broadcasters. The recent spat over carriage fees between Hathaway, an MSO, and IndiaCast Media Distribution,  a content provider (IndianTelevision.com Team 2012), shows that  intermediaries   are no  pushovers. Lack  of fair, pluralistic   reporting   may  draw   the   disfavour  of consumers  and not need regulator y policing.  With the right relaxation of regulations, barriers to entry in the TV news industry could be lowered, so that incumbents   feel  competitive   heat  from potential entrants irrespective of their market shares.

 

Promote Contestability

In certain contexts, modern regulatory economics emphasises contestability as a possible regulatory target rather than even and widely distributed market shares (Baumol 1982).  The philosophy of a  contestability- led regulatory  approach is to make barriers to entry and exit low so that incumbents are forced to strive for better quality and lower prices, no matter how concentrated the industry may be. By its very nature, the  media  industry,  especially TV,  requires  high upfront  investment. However,  competition  to TV news comes not merely from other news channels butalso from  media like radio and print.  Contestability in the news industry  as a whole  can be influenced  by measures that include  allowing  FM  radio  channels to transmit  news, and rationalising  taxes and licence fees.

Limiting   cross-media  ownership  is  preferable to controlling  shares within  a  single  media   as the economies of scale (and  bandwagon  effects  for viewers) within   a particular  media are greater than the economies of scope across media. Some countries like the UK follow the “two in three” rule whereby an entity can only be present in two out of the three media channels of TV, radio and print.

A  very  important   counterpoint   to corporate media  can  be  the  state-sponsored  TV station. However, it is usually  seen as a mouthpiece   of the government. The allocation of public money to create a truly independent news channel can be a source of great public good. Other alternatives include a public broadcasting service supported by user contributions.

There  is no  theoretical  argument  or empirical evidence on the correlation  of media plurality  and plurality of media ownership. Indian data on market shares, especially  in television,  are not reliable. In such a situation,  monitoring   the competitiveness of the  media  industry  through  traditional   economic measures  such as the Herfindahl Hirschfield Index (HHI) should be done with care.

 

Avoid Paternalism

Curtailing    a  media  company that  has grown  on the  basis  of popularity  with readers  may  not be appropriate,   and may even be regarded as an infringement of the right of free expression, unless there is evidence of an unacceptable abuse of market power  (vertical  integration  with a transmission intermediary can certainly facilitate such abuse). A certain degree of lack of plurality  may be in alignment with viewer preferences. For example, in Tamil Nadu, public opinion is split between two Dravidian parties, Dravida  Munnetra  Kazhagam (DMK)  and All India Anna  Dravida  Munnetra   Kazhagam (AIDMK),   and viewers are perhaps not unduly ruffled by the fact that the Tamil  TV channels mostly reflect the opinions of one or the other of the two parties. It may be paternalistic to impose one’s ideal of media plurality on a satisfied viewership.

 

Look Beyond Corporate Media

The regulator y approach  must  not restrict itself to media  funded  and  managed by  advertising  and,  to a lower  extent,  subscription  revenues. Such media inherently  converges to mainstream points of view – those held by the “median  viewer,” the viewer at the centre of the news spectrum, the position containing the maximum  number of viewers,  or containing viewers who  have the maximum  ability  to pay. This convergence  is easily seen in the case of news related to business, with  most media adopting  a pro-business, pro-market   stance, favouring  the  “hand  that  feeds them.”

Hence, while  attempts to make the corporate media more competitive must continue, the need to go beyond must be kept in mind. An attempt must be made to promote community-generated news through community radio and TV. Progress on community  radio has been slow and painful  in India. A landmark judgment  of the  Supreme  Court of India of 1995 stated that air waves were public  property  and must be used for the  public  good.  This  was followed by an appropriate  policy only in 2006 that allowed a wide swathe of institutions, including NGOs, educational institutions  and agricultural universities, to apply for licences (UNESCO 2011). There is still no policy on community TV even though the large cable network  in India  provides adequate transmission pipes to transmit content.

Further,  one must recognise the potential  of peer-to-peer  user-generated news. The Arab Spring is an example of the power of this phenomenon. The use of the Information  Technology (IT) Act to crack down on free expression of views is an example  of what not to do.

 

Transmission: Who Pays?

There is a feeling that the cost of the transmission of content has to be recovered in some way, and hence, one cannot be too puritanical about not charging content providers  as the net neutrality  school exhorts  one to do. Indeed,  carriage fees are standard  in broadcasting. The service of the transmitter of content  is akin to the provider of a road.  A toll to prevent congestion is acceptable on a road, so why should  it not be the same for  a provider  of connectivity services? Even if charging the content provider  violates the end-to-end design principle,  at least the end user can be charged on non-discriminator y grounds like  the quantity  of viewing time.

It is to be noted here that the toll necessary to prevent congestion and ensure efficiency may not be sufficient to fund the maintenance and upgradation of the transmission network.  Under  such circumstances, economic  theory  states  that  if the  toll good  has positive spillover effects on the economy, then it can be supported by general budgetary outlays.

The government  is already laying an optic fiber network for rural access. It can also support urban access  to broadcasting through  subsidies or by building  its own DTH  and cable network.

Transmission intermediaries  may want to differentiate  themselves from  competition  by smart traffic  management practices and innovative pricing for connectivity,  including  charges levied on content providers. However, in keeping with the end-to-end design principle,  it is better for them to recover their investment  from the  end-consumer through  non- discriminator y  usage plans,  rather  than  to actively mediate between the consumer and the content provider.  Competition   in the  content   layer  and public-private  partnership in the transport  layer of the network  are preferable to overregulation of the content layer and competition in the transport layer.

 

article 6_Table1

Conclusion

To sum up, the convergence of traditional  media and the Internet  is on its way. This convergence requires a convergence of regulatory approaches. Given the  traditional approaches taken in both  media, a synthesis of the net neutrality and the media plurality approaches is the order of the day. The following table highlights some of the key recommendations of this paper. It is time to use the power of technology to make a thousand flowers bloom – in broadcasting as well as on the Internet – and to enhance the power of community-generated media so that the need to police corporate media can be limited.

 

For further reading

Administrative  Staff College of India (ASCI) (2009). “A Study on Cross Media Ownership in India”, Ministry of Information and Broadcasting, New Delhi.

 

Baumol, W (1983). “Contestable Markets: An Uprising in the Theory of Industry Structure”, The American Economic Review, 73(3):1-15.

 

IndianTelevision.com  Team  (2012).  “DAS  Phase  II:   Indiacast- Hathway-GTPL slugfest on DAS deals”. Accessed on June 5, 2013: http://www.indiantelevision.com/headlines/y2k13/apr/apr49.php Saltzer, J H, D P Reed, and D D Clark (1984). “End-to-End Arguments in System Design”, ACM Transactions on Computer Systems, 2(4): 277– 288.

 

 

Telecom Regulatory Authority  of India (TRAI) (2009). “Recommendations on Media Ownership”, TRAI, New Delhi. Accessed on May 29, 2013: http://www.trai.gov.in/WriteReadData/ Recommendation/Documents/recom25feb09.pdf

 

United Nations Educational, Scientific and Cultural Organisation (UNESCO) (2011). “Ground Realities: Community Radio in India”, UNESCO, New Delhi. Accessed on May 29, 2013: http://maraa.in/ wp-content/uploads/2012/08/Maara_inside-pages.pdf

 

ABOUT THE AUTHORS

  • Rohit Prasad

    Rohit Prasad

    Associate Professor at Management Development Institute (MDI) Gurgaon, Member of the Subodh Kumar Committee on 2G Spectrum, Department of Telecommunications; author of Startup Sutra and co-author of Spectrum: The Conundrums of Thin Air.
Scroll To Top