Rachael Craufurd Smith and Damian Tambini*
Since 1945, UK policy on media ownership has displayed a consistent commitment to maintaining what has come to be known as ‘media plurality’ but uncertainty about how to achieve that objective without compromising media independence and freedom. The search for justiciable legal definitions of plurality and diversity favours the adoption of narrow quantitative measures, whereas normative and policy approaches see these as inadequate and favour broader definitions. Despite what is now an extensive body of empirical and theoretical research into the nature of, and rationales for, media pluralism, the search for narrow, ‘objective’ metrics has not been successful. In part this has been because the diverse concerns underlying state intervention and the distinct regulatory initiatives capable of addressing those concerns have not always been clearly distinguished. In part, because the available research has not provided sufficiently conclusive findings on which concrete initiatives can be based. Moreover, the development of media policy in countries such as the UK has been distorted by powerful media interests that have sought to drive forward a deregulatory agenda in the media field (Freedman: 2008,12-13; Dyke: 2004, 180-181; Freedland: 2011).
We start this paper by unpacking the main rationales for promoting media pluralism and suggest that, given the diverse nature of these rationales and the complexity of media markets, there is unlikely to be a single regulatory ‘silver bullet’ that can guarantee their realisation. We then map out the development of media ownership controls in the UK and consider how well the current ‘media plurality test’ stands up to evaluation in terms of its effectiveness, predictability and constitutional legitimacy. In particular, we examine Ofcom’s innovative ‘share of reference’ test and contrast this with alternative tests developed in the United States, Germany and the Netherlands. In the final sections of the paper we explore how potentially competing interests in the predictability and effectiveness of regulatory intervention can be balanced and the possibility of creating greater certainty for industry by incorporating established metrics employed in the competition field. We conclude by identifying certain lessons that can be learnt from the UK experience and suggest that there is considerable potential for further co-ordination at the European level.
2. Rationales for Market Intervention
This paper focuses on regulatory initiatives designed to ensure a plurality of independent information sources, sometimes referred to as ‘external pluralism’. By contrast, the term ‘internal pluralism’ is used to refer to those instances where different views or genres are reflected within a particular channel or media outlet (Westphal: 2002, 476). External pluralism consequently focuses on the structure of the market, internal pluralism on content, with the former seen both as a means to realising the latter and a goal in its own right. Three closely related rationales are usually cited for controlling media ownership:
a) Maintaining the integrity of the democratic process.
Where a few firms dominate the media landscape they exercise considerable control over what information is made available to the public. Dominant media companies, frequently vertically integrated with extensive content rights, can side-line, if not exclude, competitors from the market and their capacity to influence what people think they ‘know’ and consider to be important affords such companies what Baker terms ‘demogogic power’ (Baker: 2009, 65). Although information can sometimes be accessed directly from its source, few people have the time, expertise, or inclination to obtain information in this way; instead they rely on the media to research, select, and relay ‘newsworthy’ material for them. Lack of knowledge may distort or reduce political engagement (DellaVigna and Kaplan: 2007; Bernhardt, Krasa, and Polborn: 2008; George and Oberholzer-Gee: 2011, 3-4). Moreover, the capacity for individuals to be able to select their preferred source or sources of news needs to be available in the long as well as short term. As a result, state intervention may be required to ensure that the choices made by individuals at point A do not impoverish those available to individuals at point B.
Media groups often seek to leverage their real or perceived influence over public opinion further into the political domain. In the UK, leading politicians have recently revealed the extent of their contacts, formal and informal, with members of the Murdoch family and key News International (‘NI’) officials (Ball: 2011, Mulholland: 2011). These contacts raise concerns that aspects of government policy, in particular deregulatory policies favourable to major media interests, have been bartered for media support in NI newspapers (Grice: 2011). Whether or not the suspicions are well founded they undermine public trust in the democratic process. Enhanced media plurality is seen as one means of addressing concerns over the potential for undue influence wielded by powerful media owners over citizens and politicians.
b) Preventing media misrepresentation and the suppression of information.
Multiple independent sources help to ensure that media bias or misinformation stemming from particular corporate or political affiliations is exposed or deterred, thereby enhancing the quality of information made available to the public. Schwartz (2010, 48-50), Baker (2009, 656-7) and Smith (2009) all cite cases where corporate or political affiliations appear to have influenced the nature or timing of news reports. The US television network NBC is, for example, owned by General Electric, a major constructor of nuclear power plants. When an internal General Electric report was leaked indicating tests on their plants had not been adequate, NBC covered the report two months after other news outlets ran with the story. If other outlets had not covered the issue it is questionable whether NBC would have reported on it at all. Media companies have an interest in revealing unreliable or misleading journalism provided by their competitors, so that a more fragmented and competitive market can enhance the overall quality and trustworthiness of media services (Bernhardt, Krasa and Polborn: 2008, Gentzkow and Shapiro: 2006).
c) Enhancing citizens’ access to diverse information and opinions.
State intervention is sometimes justified on the basis that a variety of independent media sources is more likely to provide diverse information and opinions than a consolidated market (Beebe: 1977; Craufurd Smith: 2004, 653; Baker: 2009, 666-669; Napoli and Gillis: 2008, 8). Enhanced competition can result in a variety of editorial outlooks, with separate news teams pursuing different leads and no one organisation able to set the news agenda. The 2001 Department of Culture, Media and Sport consultation on media ownership observed ‘[d]ifferent media companies produce different styles of programming and publishing…each have a different look and feel to them. A plurality of approaches adds to the breadth and richness of our cultural experience’ (UK Gov. 2001, para. 1.7). The previous two rationales focus on the risk of distortion to the political process or coverage of specific issues as a result of powerful vested interests, whereas the focus here is on how a multiplicity of independent sources can positively enhance the provision of information, enabling individuals to make not only informed political choices but also key economic and social decisions about how to lead their lives.
Fragmentation of media ownership can thus be seen as an end in its own right, serving to dilute media power and influence, and a means to other ends, notably preventing bias and enhancing media diversity. State intervention on this basis is not, however, uncontroversial and can be challenged on both theoretical and empirical grounds. For example, concerns over undue influence can be criticised as paternalistic and incompatible with liberal notions of individual freedom and choice (Kelley and Donway: 1990), while the causal links between media ownership controls, the containment of media bias and promotion of media diversity appear complex and highly context dependent.
In relation to media power and influence, there are, as noted above, well documented examples of bias or the suppression of information resulting from political or corporate affiliations, but should we be concerned about such bias in practice? Early theories that considered media messages capable of directly influencing consumers are now largely discounted; instead, the media are seen as playing an important role in determining not only what individuals think about but also the way they think about what they think about (Iyengar: 2010). Influence thus takes place through processes of agenda setting and issue framing (ibid), and studies indicate that the media can in certain contexts influence both personal and political choices (DellaVigna and Kaplan: 2007; Bernhardt, Krasa, and Polborn: 2008; George and Oberholzer-Gee: 2011, 3-4).
Structural fragmentation of the market is a plausible technique to address concerns over potential abuses of media power, making it more difficult for one company to set the agenda or dominant frame of reference. As noted, media companies have an incentive to reveal the shortcomings of their competitors in order to enhance their own reputation for quality and reliability, so that with more firms in the market and more diverse ownership there will be less scope for media bias to go unnoticed (Bernhardt, Krasa and Polborn: 2008; Gentzkow and Shapiro: 2006; Brocas, I: 2011).
The link between structural controls and a general enhancement in content or viewpoint diversity appears more complex and problematic (see Waldfogel: 2002; Napoli and Gillis: 2008; Ho and Quinn: 2009; Baker: 2009; Schwartz: 2010; Brocas, I, et al: 2011; Waldfogel: 2011). Empirical research, backed by economic theory, suggests that where there is scope for only a few operators in a market, diversity will be enhanced by greater consolidation, with companies seeking to cover all interests rather than simply compete in the most popular fields (Steiner: 1952; George and Oberholzer-Gee: 2011, 2-3, 14 and 18). Concentrated media organisations may also be better resourced and thus able to afford costly investment in investigative journalism and original content.
On the other hand, research from the US also indicates that though cross-media consolidation can lead to an overall increase in the amount of news content relayed by the consolidating firms, overall the provision of news tends to be reduced as powerful operators crowd out alternative and possibly distinct sources (Erb: 2011, 28; and, anecdotally, Turner: 2004). Waldfogel, moreover, has found that markets with proportionally more commercial news stations in large ownership groups have ‘fewer news stations and less news variety, relative to other markets their size’ (Waldfogel: 2011, 17). And rather than enhance investment in journalism, consolidating firms may instead reduce the number of journalists in order to enhance profits or stave off losses through economies of scale (House of Lords: 2012, 42; Cooper: 2007; Hitchens: 2006, 302).
Content diversity can thus be enhanced either by consolidation or by an increase in the number of independent operators, depending on the size and revenue potential of the market. The appropriate means for realising the two plurality goals may thus be in tension, with diversity advanced by further consolidation but media power dissipated by greater structural differentiation, necessitating a choice between objectives and potential policy trade-offs (Alexander and Brown, 2007).
The position is further complicated by the fact that ownership concentration is only one among a number of variables that can influence media power, journalistic quality and content diversity (Valcke et al: 2010). Cause and effect are difficult to untangle and are highly context dependent. The type of media owner – committed family firm, for example, or major corporation with profit as its primary motive – can affect how an outlet, even a very concentrated one, operates (Cooper: 2007, 151). Similarly, consumer preferences and journalistic traditions are highly relevant. To be effective, state intervention needs to take these broader economic, corporate and cultural factors into account, suggesting that a cluster of regulatory strategies, beyond media ownership controls, is needed if we are to successfully establish a media environment where pluralism can be realised in practice.
Empirical uncertainty and the variable, complex, nature of media markets renders it difficult for states to frame robust and convincing media ownership rules. As a result, such rules are vulnerable not only to challenge on the basis of property rights and freedom of expression, but also to administrative law actions, focussing on irrationality or lack of proportionality. In the US, media corporations have had considerable success in challenging media ownership restrictions on the basis that they are arbitrary, capricious or contravene First Amendment guarantees.. In Europe, the guarantee of freedom of expression in article 10 of the European Convention on Human Rights has been successfully deployed to challenge state ownership restrictions, though states are allowed a degree of latitude in regulating media markets on pluralism grounds where this can be shown to be proportionate. More recently, the ECtHR has held that states are not merely permitted to promote plurality but are under a positive obligation to ensure ‘that the public has access through television and radio to impartial and accurate information and a range of opinion and comment, reflecting inter alia the diversity of political outlook within the country’.
At EU level state regulations designed to promote media pluralism are similarly regarded as pursuing a general interest objective that can justify restrictions on the free movement of goods or services but, to be legitimate, they must be shown to be neither directly discriminatory nor disproportionate. The importance of respect for media pluralism is explicitly recognised in article 11.2 of the EU Charter on Fundamental Rights.
States are thus under pressure to ground their media ownership rules on a firm empirical footing that appears currently elusive (Karpinnen: 2010). But our inability to draw from the existing research precise indications of the desirable level of state intervention is not a good basis for dismantling the existing structural controls. As indicated above, there is evidence that multiple sources can reduce the risk of media misinformation and abuse of power and, at least in certain contexts, result in a more varied and diverse media sector. The important democratic, social and individual interests at stake suggest that, as in the environmental field, a precautionary approach is required, with restrictions maintained until convincingly established that they are no longer needed or are ineffective.
3. Policy Rationales in the UK
The variety of rationales for domestic media ownership policy and the impact of technological and market changes can be illustrated by a brief historical overview of intervention in the UK. The first Royal Commission on the Press was asked in 1947 to examine the ‘control, management, and ownership of the newspaper and periodical press’ with the object of ‘furthering the free expression of opinion through the Press and the greatest practicable accuracy in the presentation of news’ (Royal Commission on the Press: 1949, p3).
The Commission concluded that allegations of monopoly formation and tendencies to concentration were unfounded in the UK media at that time. It did find, however, that some aspects of newspaper performance and standards were problematic and recommended the formation of a ‘General Council on the Press’, together with a range of measures to promote the press in the context of post-war austerity and a shortage of paper (p155-174). In order to counteract what were seen to be pernicious issues of proprietorial control over editorial policy, the Commission argued that a range of ownership and control structures – and in particular ownership by trusts – should be encouraged, though the report stopped short of recommending fiscal incentives to do so (p157-8). The main impact of the first Royal Commission was thus not to address structural problems of market concentration but to try to encourage improvements in press standards and journalistic ethics. Notably, it encouraged the industry to establish a self-regulatory body, the Press Council, in 1951.
Continuing concern about press concentration and a less optimistic account of concentration by the next Royal Commission in 1962 (HMSO CMD 1811) led to the establishment of a new public interest regime for newspaper mergers in the Monopolies and Mergers Act 1965. The current policy framework for media mergers can be traced to this period. The 1965 Act was then largely replicated in the Fair Trading Act of 1973, setting out a special public interest consideration, ‘in particular, the need for accurate presentation of news and free expression of opinion’, that would be applied to all significant press mergers. The Commission also recommended the formation of a specific court to adjudicate on ‘press amalgamations’ with aggregate weekly circulations exceeding 3 million copies, (HMSO CMD 1811, 1962: p118) but this proposal was never implemented. In addition, specific concentration rules limiting the number of broadcasting licences that a single individual or company could hold and cross media ownership thresholds were established in the 1965 Monopolies and Mergers Act. These were immediately tested by the referral, and ultimate approval, of the proposed purchase of The Times by Lord Thomson in 1966, who already owned The Sunday Times.
There followed a period in which newspaper sales and advertising revenues declined, marked by a number of controversial mergers, such as the 1981 purchase of the Observer newspaper by Lonrho and The Times and Sunday Times by News Corporation. The latter was allowed by the Department of Trade and Industry without reference to the Monopolies and Mergers Commission on the basis that the papers were failing, though structural changes were required, notably the introduction of independent directors, intended to protect the papers’ editorial independence (DTI: 2004, para 5.60). A further Royal Commission 1974-77 (Third Royal Commission on the Press: 1974-77) was followed by a review of media ownership by the Department for National Heritage (‘DNH’) in 1995, this time with a focus on the evolving broadcasting environment and the impact of new cable and satellite services (DNH, 1995). DNH set out two major issues for consideration:
‘- whether competition law on its own could be relied upon to provide the degree of protection against abuse of position in the media industry, which Parliamentary and public opinion expect, and
– whether license conditions and positive programming requirements could be relied upon to deliver sufficient diversity, regardless of ownership’ (5.3, p19).
DNH concluded that reliance could not simply be placed on general competition law, whose primary focus was safeguarding ‘the process of competition and the interests of the consumers as purchasers of goods and services’ (5.8, p17). In particular, it noted that ‘the media do not supply a simple product or service – they are the means of mass communication through which society develops opinions’ (5.8, p17). Rather than wholesale deregulation, targeted relaxation of the existing ownership limits was proposed, primarily of benefit to satellite broadcasters and the press (Hitchens: 2006, 99). These were given effect to in the Broadcasting Act 1996.
The 1995 report also tabled longer term proposals for the establishment of a new regulator with responsibility to ‘govern media ownership’ (ibid, p22) and set thresholds for merger referral. In recognition of the increasingly convergent media environment an ‘exchange rate’ expressing the relative influence or market power of different media would be developed. The aim was to establish a radical new framework for media ownership, with new methodologies for media market definition; recognition of the different degrees of influence of the various media; ownership thresholds triggering a public interest evaluation; and new powers for the regulator to prevent acquisitions or require divestment. These ambitious proposals, however, were not implemented, due to the change of government and ongoing uncertainties about the proposed methodologies, notably as to the viability of a depoliticised ‘exchange rate’ for evaluating media influence.
At present, the commercial media landscape in the UK is governed by the 2003 Communications Act. The New Labour Government elected in 1997, continued to emphasise the importance of plurality for the democratic process but concluded that technological developments had opened the way for new market entrants and were keen to encourage investment into the industry and to support internationally competitive domestic players. As a result, many of the rules preventing the accumulation of broadcasting licences were removed, leaving only the prohibitions on cross media concentration at the local level and between the press and the holder of the Channel 3 television licence at the national level. Instead, increasing reliance was to be placed on the operation of competition law to protect media plurality, coupled with remaining rules preventing certain types of owner, for instance political bodies and advertising agencies, from obtaining broadcasting licences; content regulation, notably impartiality requirements for the broadcast sector; and state subsidies, particularly for public service broadcasting.
Contrary to the conclusions of the DNH in 1995, the government had initially considered that competition law could be relied on to prevent undue media consolidation at the national level. During parliamentary consideration of the Communications Bill in 2002, however, the joint committee on pre-legislative scrutiny (‘Puttnam Committee’) took issue with this conclusion and argued strongly for the Competition Commission to be given new powers to consider not only the potential for a merger to result in a substantial lessening of competition, the test applied in competition law, but also its implications for media plurality (Joint Committee Report: 2002). This led to the retrospective amendment of the Enterprise Act 2002 to incorporate a number of distinct ‘media public interest considerations’ (‘MPIs’), discussed in more detail in the following section, which relate to different aspects of media pluralism. These replaced the press public interest test in the 1973 Fair Trading Act and added a further set of tests for mergers involving broadcasting companies.
Government guidance around the passage of the Communications Act 2003 confirms that these provisions were intended to perform two distinct functions, which broadly relate to the rationales identified in section 2 above: firstly, to prevent undue accumulation of power and influence, and, secondly, to preserve media diversity. Thus, the relevant Government minister, Lord McIntosh, stated that ‘[m]edia plurality is important for a healthy and informed democratic society. The underlying principle is that it would be dangerous for any person to control too much of the media because of his or her ability to influence opinions and set the political agenda’ (Ofcom: 2010b, para 1.5). Baroness Blackstone, introducing the 2003 Act in the Lords stated that one of its main purposes was ‘to ensure the existence of a range of media voices, safeguarding the vibrancy of democratic debate’ (quoted in Enders, 2010 at 16). The public interest considerations identified in the Enterprise Act 2002 explicitly include references to a ‘sufficient plurality of views’ in newspapers and a ‘wide range of broadcasting…calculated to appeal to a wide variety of tastes and interests’ (s. 58, see further section 3 below). Subsequent DTI guidance also noted that the provisions were intended to ‘ensure a minimum level of plurality’ (Ofcom 2010:5).
The importance of the plurality test is underscored by the passage of the Media Ownership (Radio and Cross-media) Order 2011, which removed all remaining restrictions on media accumulations at the local level. Although Ofcom, the UK’s main independent commercial communications regulator had concluded in 2009 that combined ownership of a Channel 3 regional television licence, local analogue radio licence and the main local newspaper all operating in the same area could confer ‘too much control over the local news agenda’ to one person or company, it later concluded, after the matter had been referred back to it by the Coalition Government, that even concentrations of this scope might not be judged unduly detrimental. This, it noted, was ultimately a policy decision for Government to take (Ofcom: 2010c especially at 3.13-3.16). In reaching this conclusion, Ofcom emphasised not only the presence of the BBC as an alternative source of news and the Government’s plans for new local television stations but also the ‘media plurality test’. Review under the Enterprise Act 2002 is now, therefore, the key mechanism by which concentration of media ownership in the UK can be controlled on plurality grounds.
4. Key Characteristics of the UK Media Plurality Test
The Enterprise Act 2002 enables the Secretary of State to issue an ‘intervention notice’ when he or she considers that an impending media merger could raise media pluralism concerns. This differs from the Fair Trading Act 1973, which required the Secretary of State to give written consent to all mergers with a combined daily circulation in excess of 500,000 copies (s.57-62). Application of the Enterprise Act 2002 requires the company being taken over to have a turnover in excess of £70 million or for the merger to result in a 25% share of supply of goods or services across the UK or a substantial part of it (s.23). In addition a special intervention notice can be issued where the merger creates or enhances a 25% share of the supply of ‘any description’ of broadcasting services or newspapers in a ‘substantial part’ of the UK, or for one of the merging parties to already hold such a share (s.59, DTI: 2004, para.3.7). In addition, sections 67-8 of the Act enable a ‘European intervention notice’ to be issued where, owing to the scale and European dimension of the merger, the competition questions are to be addressed by the European Commission. In this context the EU and the UK engage in complementary but distinct investigations within the terms of article 21(4) of the EU Merger Regulation, with the Secretary of State remaining competent to consider the plurality issues.
Although the plurality test was seen as an integral part of the 2003 deregulatory package, recent events have illustrated that tests of this nature can potentially lead to intervention in contexts that would previously have been considered unproblematic, thereby increasing rather than reducing regulatory oversight (Competition Commission: 2007, paras.5.30 and 8.8). The intervention notice triggers an initial investigation by the OFT into any competition aspects and by Ofcom into the stated media plurality concerns. On the basis of the OFT and Ofcom reports, together with any representations made as a result of further consultations, the Secretary of State will decide whether to refer the matter to the Competition Commission for a more detailed examination; to accept behavioural or other modifications designed to address any concerns raised; or to allow the merger to proceed. Where the matter is referred on to the Competition Commission, the Secretary of State will consider the Commission’s report and any further representations before deciding whether or not to allow the merger and, if to allow it, on what terms. He or she is under no obligation to follow the Competition Commission’s advice on the plurality issues, though is so bound regarding any competition points. An aggrieved party can seek review of the Secretary of State’s decision before the Competition Appeal Tribunal (‘CAT’) and rulings of the CAT can be appealed, provided leave is granted, to the Court of Appeal in England or Court of Session in Scotland.
The public interest criteria are set out in section 58 of the Act and differ depending on whether the merger involves just newspaper publishers or broadcasters. In relation to newspaper mergers, the MPI considerations are expressed as:
58(2A) The need for:
(a) accurate presentation of news; and
(b) free expression of opinion;
58(2B) The need for, to the extent that it is reasonable and practicable, a sufficient plurality of views in newspapers in each market for newspapers in the United Kingdom or a part of the United Kingdom…
In relation to broadcasting and newspaper/broadcast cross-media mergers the considerations are:
(a) the need, in relation to every different audience in the United Kingdom or in a particular area or locality of the United Kingdom, for there to be a sufficient plurality of persons with control of the media enterprises serving that audience;
(b) the need for the availability throughout the United Kingdom of a wide range of broadcasting which (taken as a whole) is both of high quality and calculated to appeal to a wide variety of tastes and interests; and
(c) the need for persons carrying on media enterprises, and for those with control of such enterprises, to have a genuine commitment to the attainment in relation to broadcasting of the standards objectives set out in section 319 of the Communications Act 2003.
The specified concerns broadly reflect the objectives identified in section 2 above. These, however, are afforded patchy and, in some cases rather oblique, recognition. Thus, source plurality is identified as a concern solely in relation to broadcast or cross-media mergers, whereas the goal of promoting viewpoint diversity finds reflection in both sets of provisions. This suggests that the reference to source plurality was intended to address undue media power or influence rather than diversity issues, reflecting the concerns expressed by Lord McIntosh during passage of the Communications Act (DTI: 2004, para. 7.7). The concern to prevent bias or misinformation is reflected in both the newspaper and broadcasting/cross-media contexts by virtue of sections 58(2B)(a) and 58(2C)(c).
The adoption in 2003 by the UK of a ‘flexible’ case by case approach to addressing media plurality concerns can be contrasted with the then position in the US, where the Federal Communications Commission (‘FCC’) was instead establishing exact limits to the number of media interests that could be accumulated in specific markets. The FCC’s preference for a ‘bright line’ approach was based on its conclusion that ‘[a]ny benefit to precision of a case-by-case review is outweighed, in our view, by the harm caused by a lack of regulatory certainty to the affected firms and to the capital markets that fund the growth and innovation in the media industry’ (Goldfarb: 2004, n.34). The US concerns over certainty seem justified in light of subsequent UK experience with the open-ended MPI criteria discussed below.
5. The Media Plurality Test in Operation
At time of writing, only two intervention notices on media plurality grounds have been issued. The first concerned the 2006 purchase by satellite broadcaster British Sky Broadcasting (‘Sky’) of a 17.9% shareholding in terrestrial television broadcaster ITV (the ‘Sky/ITV’ case). The case highlighted the procedural complexities of the system, with Ofcom, the Competition Commission, the Secretary of State, the CAT and, ultimately, the Court of Appeal all considering the plurality issue. The whole procedure took more than three years to conclude (Craufurd Smith: 2009; Arnott: 2010). It also revealed the degree of interpretative latitude afforded by the Enterprise Act: the intervention notice in this case referred to section 58(2C) of the Act, which focuses on the need for a ‘sufficient plurality of owners’, but the section provides no guidance as to what might constitute a ‘sufficient plurality’ or the basis on which this should be determined. With little, if any, precedent to go on, divergent views were inevitable and in the Sky/ITV case the conclusions of Ofcom and the CAT, on the one hand, differed markedly from those of the Competition Commission, the Secretary of State, and the Court of Appeal, on the other. Though the merger was ultimately found not to threaten media plurality, divestiture by Sky of the bulk of the ITV shares was ordered on competition grounds.
The second case involved the bid by News Corporation (‘News Corp’) to purchase 61% of the shares in Sky, alongside its existing 39% holding, thereby gaining complete control of the company (the ‘NewsCorp/Sky’ case). News Corp, through its subsidiary News International, currently publishes The Sun, The Times and The Sunday Times newspapers in the UK and, until its recent closure as a result of the phone hacking scandal, the News of the World – now replaced by the Sun on Sunday. The intervention notice in this case was similarly based on section 58(2C) of the 2002 Enterprise Act but evaluation of the merger was itself engulfed by the phone hacking dispute and in July 2011 News Corp withdrew its bid. Although there had been no definitive ruling in the case at this point, Ofcom had completed its provisional report on the proposed acquisition and the Secretary of State was in the process of examining whether News Corp’s proposed spin-off of Sky’s dedicated news service, Sky News, into a separate company would address the plurality concerns raised by Ofcom. News Corp would have held a 39% holding in the separate news provider, the same interest it currently holds in Sky, enabling it to argue that the proposals would have no impact on the plurality of available news sources.
The News Corp/Sky case illustrated, if illustration was needed, that competition law review, though potentially a useful supplement to media specific ownership controls, is not a complete substitute for them. The merger was referred to the European Commission, which is competent under the European Union Merger Regulation to consider mergers over a certain threshold that have a European dimension. When assessing the News Corp/Sky merger the Commission considered whether the merged entity would be able to restrict competitors’ access to premium films and advertising space or undermine competing newspapers by offering mixed bundles of print and audiovisual content at attractive rates. All these activities would make it difficult for alternative companies to survive in a competitive market and could consequently reduce the number of alternative sources available to consumers. The European Commission concluded that the merger would not enable the new company to restrict competition in this way.
The European Commission did not, however, consider the impact of the merger on the market for news in general, investment in independent journalism, or the range and content of programming, in particular news programming, provided by the merged firm itself. From the perspective of competition law the consolidation of news offices can lead to positive market efficiencies but from the citizens’ perspective this may threaten continued access to diverse information. Competition law focuses on the exogenous effects of a merger on competing firms rather than the endogenous effects of the merger on the merging firms’ own services. Moreover, article 21(4) of the Merger Regulation makes it clear that the regulation does not encompass issues relating to media plurality, which are explicitly left for Member State determination.
It is beyond the scope of this article to explore the MPI investigations in the Sky/ITV and News Corp/Sky cases fully but two aspects merit particular attention: the definition of the relevant field and how sufficient plurality was assessed. In relation to the relevant field, it is necessary to consider both content and medium. On content, all of the bodies involved in assessing the mergers followed Ofcom’s lead and focused on the provision of news and current affairs, more particularly national news (Ofcom: 2010b, 6-7). This was not an inevitable conclusion and countries such as Germany look to the audience share of television channels generally when assessing media concentration. Ofcom argued that news was considered by consumers to be of the greatest ‘personal and social importance’, was important for the formation of public opinion on matters of national significance and was, consequently, central to concerns over the operation of the democratic process (Ofcom: 2010b, 6). As a result it focused on the main providers of UK national news services and did not take into account minority or international news services such as Al Jazeera and CNN.
In relation to the relevant medium, it was recognised in both cases that individuals increasingly draw on a wide range of available media for their news – broadcast television and radio services, printed newspapers and internet services – all needed to be taken into account when assessing the impact of a merger (Ofcom: 2010, para.2.17-22). In both cases, however, the ongoing dominance of the mainstream media was noted, particularly television, which, in the News Corp/Sky case Ofcom found to be the main source of news for around 73% of the UK population (Ofcom: 2010b, para.4.9, see also Competition Commission: 2007, para. 5.44-5.45).
In Sky/ITV Ofcom engaged in two separate examinations, one on the impact of the merger on the cross-media market for national news and one on the television market for national news (Ofcom, 2007: 1.11). In NewsCorp/Sky it focused on the cross-media audience for news and current affairs. In neither case did Ofcom consider the field of publishing more generally or the role of search engines and other intermediaries. Although the latter play an increasingly important role in the online world in determining what content individuals access in practice they are primarily aggregators of third party content not producers in their own right.
This brings us to the second question: how was ‘sufficiency of plurality’ assessed? In Sky/ITV although the Competition Commission noted that the UK media ownership controls had in part been introduced to address concerns over undue power and influence it concluded that plurality of control was important because of its potential impact on the range of information made available to the public (Competition Commission: 2007, para.5.10). The Commission’s focus on content diversity meant that it was willing to accept a reduction in source diversity where it could be shown that content diversity, and in particular the diversity of content provided by the merging firms, was not affected. It thus looked ‘qualitatively’ at the sufficiency of providers and concluded that the merger would not enhance BSkyB’s control over ITV to a degree that could affect ITV’s provision of news (Competition Commission: 2007, paras.5.15 and 5.73-5.74).
In contrast, the CAT regarded section 58(2C)(a) of the Enterprise Act 2002 to be concerned with undue influence over public opinion and the news agenda, the first of the rationales in section 2. It therefore emphasised the importance of structural media plurality and observed that such plurality, once lost, is difficult, if not impossible, to restore (CAT: 2008, para.262). The dispute thus illustrated the way in which underlying objectives influence how ownership structures are assessed. In particular, the CAT regarded section 58A(5) of the Enterprise Act 2002, which requires an acquisition to be deemed to reduce the number of firms operating in the market by one, to have been introduced on legitimate ‘precautionary’ grounds, and, as a result, Sky and ITV had to be treated as under single control. Although both qualitative and quantitative assessments were held to be relevant when assessing the impact of the merger on the overall plurality of sources, section 58A(5) precluded consideration of the extent to which Sky could control ITV’s news content and the internal plurality or range of views provided by the merged companies (Competition Appeal Tribunal: 2008, para. 265 and, for further discussion Arnott: 2010). Ultimately, it was the view of the Competition Commission that was endorsed by the Court of Appeal, which noted that plurality could ‘carry an implication of range and variety as well [as multiple sources]’.
The Secretary of State’s notice in the subsequent case relating to News Corp’s proposed acquisition of Sky similarly referred to section 58(2C)(a) of the Enterprise Act 2002. In line with DTI guidance and the view of the CAT, Ofcom emphasised the importance of source diversity for ‘a healthy and informed democratic society’ and the need to prevent undue accumulations of power (Ofcom: 2010b, paras.1.5 and 1.11). It concluded that a simple head count of companies or news providers was relevant but not sufficient because not all outlets are similarly able to influence public opinion and set the news agenda (Ofcom: 2010b, paras.1.20 and 5.11). In the US, a failure to recognise precisely this point led to the striking down of the FCC’s ‘media diversity index’, used to establish media concentration thresholds, in the first Prometheus Broadcasting case.
Ofcom also made reference to the Competition Commission’s ‘qualitative’ assessment of a sufficient plurality of controllers and considered the degree of internal plurality that would be retained within the merged group (Ofcom: 2010b, paras 1.14 and 1.36). Given the complete nature of the merger Ofcom came to the provisional view that the editorial independence of the Sky News service would not be sufficiently guaranteed (Ofcom: 2010b, para 1.39). Although the existence of impartiality requirements constrained to some extent owner influence over broadcast news content, there remained, in Ofcom’s view, residual scope to determine the news agenda. Impartiality on its own did not ensure that news coverage would be sufficiently diverse (Ofcom: 2010b, para. 1.42).
In the Sky/ITV case the two companies were primarily concerned with television broadcasting but in the News Corp/Sky case both print and television services were under consideration. How can one assess the impact of a company’s cross-media holdings on public opinion in this context? As discussed further in section 6 below, measurements such as the time individuals spend accessing sources, audience share, revenue and reach can be computed for each firm’s different media holdings and added to give a global figure (Robinson:1995). But this requires a consistent approach on the part of the various monitoring bodies, something that is not presently the case in the UK. Radio statistics, for example, are based on a self-reported diary, whereas television audience measurement is based on remote monitoring of a large sample of the audience, recording when a television channel is switched on, but not who is watching it. Only television statistics provide specific information on news audiences. Print data record readership for each issue regardless of time spent and there is no breakdown of the time spent reading news items as opposed to entertainment features or advertisements (Ofcom 2010b: para 1.23, appendix 1). There is currently no standard industry measure for audience share of news websites on the Internet. As a result, Ofcom’s findings regarding reach and news consumption obtained using the existing data need to be treated with caution (Ofcom: 2010b, paras.1.23, 5.24-30, figure 1).
But even if comparable statistics were to be readily available, combining audience share, time-use, reach or revenue figures for all media to assess potential influence will remain a problematic exercise because of the possibility that individuals relate differently to different media. Television’s ability to combine images and sound, for example, is widely considered to make it a particularly influential medium, so that five minutes watching a television news bulletin may not have the same impact as five minutes reading a national newspaper (Ofcom: 2010b, para.5.29-30, for critique McQuail: 2010, 125-126). Moreover, variations in audience size for the different media means that simply combining a company’s audience shares for each medium may give a misleading indication of its influence: a 30% audience share of the radio market may afford very different scope for influence than a 30% share in the larger television market.
To address these concerns, certain countries give each medium a specific weight to reflect its presumed capacity to influence public opinion. In Germany, for example, an audience share limit is employed. Companies that attract more than 30% of the television audience are presumed to exert too great an influence on public opinion. But even where this threshold is not met a company will be found to command undue influence if its combined media services are deemed to have an influence equivalent to a 30% share of the television market. When the Axel Springer publishing group attempted in 2006 to take over Pro7/Sat1 Media AG, Germany’s biggest private television company, the television audience share of the merged company would have been 22% and thus below the threshold. The German Media Concentration Commission, ‘KEK’, however, applied a weighting coefficient of 2/3 so that Springer’s 26% of the daily print market was held to equate to a 17% share of television viewers. This, together with Pro7/Sat 1’s television share, brought the merged company over the concentration limit (Just: 2009, 110-112; Czepek and Klinger: 2010, 840; KEK, 2007).
KEK was willing to include Springer’s press interests because of the press’ capacity to influence public opinion and comparability in significant respects to national television, taking into account the medium’s suggestive power, breadth of impact and topicality (Just: 2009, 110-112; Scheuer: 2006). Other holdings considered potentially relevant were Springer’s interests in programme guides, public interests magazines, radio and the internet, each afforded a distinct conversion ratio (ibid). The decision was overruled by the Bavarian Administrative Court in February 2012, in part on the basis that Pro7/Sat 1’s television audience share was lower than the 25% threshold at which additional interests could be taken into account (Yliniva-Hoffmann: 2012).
Weighting is, however, controversial, not least because specific products within media sectors are more or less influential, and specific individuals attach very different weights to particular print or audiovisual services (McQuail: 2010, 73-74, 126; Ofcom: 2010b, paras.4.85-4.86). Without further research into how individuals receive and process information from the various media sectors, attempts to weight different media sectors potentially add to, rather than resolve, the existing methodological problems (Robinson: 1995). Such problems are compounded by rapid change in the media consumption habits in the current period.
In the UK, weighting of this kind was envisaged by the DNH in its 1995 report (p.23), noted above, but was not pursued, in part because of the difficulty of establishing convincing conversion rates. To address these problems, Ofcom developed a novel ‘share of reference’ approach. This involved a survey of 2,018 adults, who were asked to indicate from a list which news sources they used on a regular basis (weekly for daily sources, monthly for weekly sources). They were also asked to indicate which of these they regarded as their ‘main source of news’ (Ofcom: 2010b, 96). The sources ranged across all platforms and included an ‘other’ category, which identified a wide range of international and regional media as well as online search engines and websites. Ofcom used the regular use data to determine for each provider or title, firstly, its overall ‘influence’ by calculating the number of references it received as a proportion of the total number made, and, secondly, its reach, by calculating what proportion of those surveyed accessed it on a weekly basis (Ofcom: 2010b, Annex 1).
Ofcom’s approach consequently enabled it to adopt a cross-platform analysis, without artificial weighting, based on what individuals actually said they consumed. The approach can accommodate new and evolving forms of media use and enabled Ofcom to focus on its target content, news, though other genres could similarly be identified. Ofcom also configured its share of reference data from the perspective of wholesale and retail provision (Ofcom: 2010b, para.4.14). This is an important distinction in that many news channels and services rely on information provided wholesale by third parties and thus contribute little, in qualitative terms, to media pluralism. The survey revealed that the BBC, on both the wholesale and retail analysis, had the largest share of reference (37%), with a second group of significant providers, ITN, News Corp and Sky, all with around 10% on the wholesale analysis, with Sky falling back to 5% on the retail analysis. There then followed a third group of seven or eight providers with shares of 1-5%, with a final numerically large group of providers with very small shares of reference.
Reviewing this data, Ofcom provisionally concluded that a combination of the second and fourth largest news providers could ‘be expected to operate against the public interest’ (Ofcom: 2010b, para. 1.57). In particular, the merged company, though commanding a lesser share of reference than the BBC, would still have had a significantly larger share (22% wholesale/17% retail) than the next provider, ITV, (12% wholesale/9% retail). The creation of one company with, in relative terms, a much greater capacity than any other commercial organisation to influence the public was central to Ofcom’s conclusion. In this respect, Ofcom properly recognised the distinct nature of the media diversity and media influence rationales for intervention. Attempts by the Secretary of State to reach an agreement with News Corporation over the acquisition and the withdrawal of the bid in 2011 in light of the phone hacking revelations meant that Ofcom’s proposed share of reference methodology did not receive further formal consideration.
6. Options for Reform
From both citizen and industry perspectives the current UK media public interest test is patently unsatisfactory. This has been acknowledged by both the present Secretary of State for Culture, Jeremy Hunt, and the Prime Minister, David Cameron (Hunt: 2011b; Mulholland: 2011). The case-by-case approach, in line with FCC predictions, has created considerable uncertainty, not least because the few cases investigated to date leave many issues unresolved. Below, we set out key considerations that should inform the drafting of any future media pluralism regime. We also identify the main options for reform, noting those instances where further research is needed to strengthen the case for regulatory intervention.
a) The need for clear articulation of the objectives underpinning intervention
As indicated above, the Enterprise Act 2002 identifies a number of different plurality concerns for newspaper and cross-media mergers respectively. Not all these concerns need be addressed in a specific case and regulatory bodies have differed both in their interpretation of these concerns and in their willingness to prioritise certain interests over others. The present rules consequently have the potential to operate in a random and partial manner, undermining public confidence in their application. To address this, a clear and comprehensive statement of the various concerns underpinning ownership controls should be set out in governing legislation.
We suggest that these should include, as a minimum, ‘the need, in relation to every different audience in the United Kingdom or in a particular area or locality of the United Kingdom, for there to be a sufficient plurality of persons with control of the media enterprises serving that audience’ (as in section 58(2C) of the Enterprise Act); the need for the expression of a sufficient plurality of views and range of content to serve those different audiences; and a commitment to accuracy, particularly in the presentation of news, and freedom of expression. Each consideration should be taken into account when reviewing potential mergers or the state of media markets, whether these concern ownership of companies providing audiovisual or text services or a combination of both. Concerns, in the broadcasting context, over quality provision and compliance with the Ofcom Code in section 58(2C) of the Enterprise Act 2002 can be addressed more appropriately through Ofcom’s regulatory powers. Consideration could, however, be given to extending the ‘fit and proper person’ test that applies to the award of broadcasting licences in section 3(3) of the Broadcasting Act 1990 to ownership of UK media interests more generally.
b) The need to match regulatory strategies to identified public interest objectives
The reason it is important to clearly identify the basis for intervention is that certain regulatory strategies are suited to address particular concerns but not others (Baker: 2007, 83-84; Alexander and Brown, 2007). There is, for example, evidence that the existence of multiple independent sources can correct or prevent bias and may promote viewpoint diversity but multiple sources offer only limited protection against undue media influence. This is because we need to distinguish between available and consumed diversity: even in a market with many sources, one or two operators can still command most of the audience and thus determine what the majority of people read or watch.
It is consequently in keeping with the US focus on viewpoint diversity that the main metric used by the FCC to regulate media ownership is the number of independent ‘voices’ or sources in specified media markets (FCC order). But the UK also seeks to prevent undue media power and influence, which requires consideration not only of the number of outlets but also their individual impact. For this, considerations such as a company’s share of audience, time-spent consuming specific services, revenue or reach appear more relevant, together with share of ‘regular’ or ‘main source’ attributions, along the lines developed by Ofcom in its NewsCorp/Sky report.
In relation to opinion-forming power, subjective assessments of influence obtained through consumer surveys are potentially the most relevant form of measurement. But the process by which the media influence consumers remains controversial and individuals inevitably interpret media messages in different ways (McQuail: 2010, 73-4). Moreover, consumers may have only limited understanding of how they have in fact been influenced by media exposure: they may, for example, over-emphasise the impact of sources with which they agree and downplay those with which they disagree, even though the latter help to shape how they categorise issues or see the world. Napoli has identified two criteria that may provide the basis for more objective measurement, namely recall and attitude change (Napoli: 2011, 91). Audience recall can be measured relatively easily, and is already being recorded by some companies for advertising purposes, but assessing how individual attitudes or behaviour has been affected by media exposure, particularly outside the advertising context where a change in purchasing behaviour can be identified, remains difficult (Napoli: 2011, 108). Even when dealing with clearly identifiable decisions, such as voting preferences, it is difficult to disentangle the effect of media exposure from other sources of influence.
Ofcom’s ‘main source’ survey could be seen as inviting subjective assessments of influence on these lines, though the term ‘main source’ is open to divergent interpretations. Nor did the survey allow individuals to specify more than one ‘main’ source or indicate the relative importance they ascribed to the various sources they used. Individuals could instead be asked to identify all sources they consider to be influential and to evaluate them according to a scale of importance: very influential, quite influential etc (Morisi: 2012). Although it would be possible to combine personal assessments of influence, together with investigations into recall and attitude modification, data of this kind needs to be treated with caution given our present state of knowledge and used, if at all, to supplement other findings, rather than as an autonomous basis for regulation.
Revenue-based measurements are also, if more tangentially, linked to influence, in that advertising income reflects advertiser assessments of the value of reaching particular audiences, either because of their scale or demographic profile. The advertiser is prepared to pay for advertising space because it considers the audience susceptible to its message. But if we start from the premise that each citizen has an equal voice in the democratic process it is not clear why ownership restrictions should be influenced by specific audience characteristics, such as wealth, that affect advertising revenue. If, on the other hand, we consider certain individuals to be socially and politically more influential than others, concentration of ownership in the sources accessed by this group could be cause for particular concern. But we then need to have a much clearer idea of who these individuals are and the nature of their influence: reliance on the advertising preferences of major corporations does not appear a particularly compelling basis for their identification. Moreover, the value to the advertiser of access to a particular audience may bear little relation to the influence of the programme or column in which the advertisement is inserted on its audience, though recent evidence suggests that the level of audience engagement with the surrounding content positively affects receptivity to the embedded advertising (Napoli: 2011, 98). Revenue can equally be derived from subscription charges, commercial ventures, or state funding, making it difficult to draw comparisons across different outlets. Revenue-based measures thus appear poor indicators of influence.
Nor can audience share, time-spent, reach, even Ofcom’s regular source metric, which all relate to varying degrees to exposure, be directly correlated to influence. This is because, as noted above, influence can be as much a product of quality or reputation as scale of exposure (Competition Commission: 2007, para.5.41; Robinson: 61). The various forms of measurement traditionally used to fix ownership limits thus appear inapt or, at best, indirect mechanisms for assessing influence (Iosifidis: 2010, 17). But inaction in the face of uncertainty is, as noted above, here an unattractive option. Though a simple correlation between exposure and influence seems improbable, the ability to command an audience remains a plausible indicator of media power and influence, in that exposure to a particular source affords that source at least the potential to influence those who access it. Furthermore, sources that command a large audience may be believed by politicians or other decision-makers to be influential, regardless of their actual impact, and thus exert an influence on the behaviour of these groups.
Certain dimension of ‘exposure’ are also open to objective monitoring, providing a more robust basis for intervention than personal evaluations of influence, which, as noted, call for a high degree of media literacy and self-awareness. Specific dimensions of exposure that are suggestive of higher or lower intensities of audience engagement include: the total time spent accessing a particular source (overall exposure), the regularity with which the source is accessed (loyalty), and the duration of each visit to that source (attentiveness) (Napoli: 2011, chapter 3). The influence of a particular source may also be affected by whether the person accessing it makes use of alternative sources (multi-sourcing). Factoring these dimensions into future investigations of exposure should enable us to gain a better understanding of potential influence.
A time-based methodology designed to assess opinion-forming power was proposed by the Dutch Media Authority in 2011, which focused on overall exposure. The Dutch model involved a survey of 1,195 Dutch adults over 13 years in age (Commissariaat Voor De Media: 2011, 86-91). Those surveyed were asked what news titles they had used the previous day and how long they had spent accessing them. The Authority then computed the overall reach of news services according to owner and what they termed the ‘news market share’. This was the percentage of the total time spent accessing news titles that could be ascribed to news titles from a particular supplier. Time-spent accessing a source provides a ‘common currency’, enabling exposure figures for the different media to be combined, but weighting will remain an issue for time-based measures if some media are considered more influential than others. The Dutch model did not assess the loyalty, attentiveness or multi-sourcing aspects of exposure.
Ofcom’s share of reference approach similarly relied on individual survey data but focused instead on the regularity of exposure (loyalty). Ofcom also investigated reach and considered share of overall news consumption to be a ‘useful complementary measure’ (Ofcom: 2010b, para. 5.45, appendix 1). Regular access certainly suggests that the user considers the source to be useful for his or her information needs but there is unlikely to be a simple correlation between regular access and influence. A source accessed infrequently but for extended periods could well be more influential than sources accessed regularly but for short durations. Moreover, the Ofcom survey asked individuals to identify those sources they accessed ‘at least once a week’, so that sources accessed once a week were afforded the same weight as those accessed daily, leading to the impact of certain sources being potentially understated. Ofcom was also unsure how multiple-sourcing affected influence and did not factor this into its evaluation (Ofcom: 2010b, paras. 5.115-5.116 and for critique News Corporation: 2011, 6). These observations suggest that Ofcom’s share of reference approach, even in a modified form, is insufficient on its own as a proxy measure for influence.
Reach tells us how wide-spread exposure to a particular source is across all users, while audience share figures indicate what proportion of the total audience select particular services, genres or programmes. But because programmes vary in length, time-based share of consumption figures give a better indication of exposure to news or other specific types of content than audience share figures.
Of these various measures, overall exposure to a specific source appears the most plausible single indicator of potential power and influence but cumulative application of tests that address other dimensions of exposure (loyalty, attentiveness, multi-sourcing) should provide a more sophisticated understanding of the potential impact of media exposure and thus strengthen regulatory assessments of media plurality (Robinson: 1995, 71). A single survey could deal with these various indirect indicators of engagement/influence, as well as more direct personal assessments (Commissariaat Voor De Media: 2011, 90, see also Napoli: 2011, 94-101). Such a survey needs, however, to be informed by a proper understanding of what these various factors tell us about media influence and, importantly, how they interrelate. Simply multiplying metrics without further research, could well compound, rather than reduce, existing uncertainties.
c) Delimiting the market for information
Establishing appropriate metrics is only half the battle: a number of important policy decisions regarding the products covered by, and geographic scope of, the investigation will still have to be taken. In relation to the product market, key questions are whether consideration should be given to concentration in each media sector separately, as well as across all media platforms, and whether the focus should solely be on mainstream news and current affairs services.
The answer to the first of these questions is rendered more complex by media convergence. The principle of platform neutrality suggests that we should not distinguish goods or services simply on their mode of transmission but evidence that different types of media provide the consumer with rather different sets of information and employ different presentational styles, will, however, strengthen the argument for imposing concentration limits in each of the three main media sectors (video, audio, text) not simply across a combined media market. The impact of concentration in one media sector on potential influence or content diversity cannot be fully offset by a range of independent sources in another.
In relation to content, there is, as noted above, controversy over whether controls should be established in relation to the provision of news services or content more generally and, if news, whether this should extend to foreign as well as national news (Arnott: 2010, 266-270). Some countries, such as Germany, when assessing the market share of each outlet, combine television and television-like services but do not break down the data further according to genre (Czepek and Klinger: 2010, 840; KEK, 2007). This can be justified on the basis that although news is centrally important for the democratic process and plays a key role in opinion formation other genres also provide information and can be socially and culturally influential. An individual’s understanding of his or her community and the world at large may be gleaned as much from scientific, documentary, history, even drama and entertainment programmes, as from news. The popular US television series ‘24’, which repeatedly raised the question of the legitimacy of torture in the face of imminent terrorist attacks, led, for example, the US military to express concerns that the programme could be influencing the attitudes, if not behaviour, of young US troops (Buncome: 2007). In light of these observations it has been suggested that specific attention should be paid to the impact of a transaction on ‘cultural provision serving any group of users or viewers that might be significantly affected by it’ (CCMR: 2011, 13).
But news and current affairs programmes do appear particularly relevant for opinion formation in that, as the Dutch Media Authority has noted, ‘the news media’s ability to transfer issues of importance from their own media agendas to the public agenda makes them very powerful, and thus essential in assessing concentration of power in the process of opinion formation’ (Commissariaat Voor De Media: 2011, 87). There are also practical advantages to focusing specifically on news. The FCC, for example, has concluded that ‘news programming is more easily measured than other types of content containing viewpoints…[and] relates most directly to the Commission’s core policy objective of facilitating robust democratic discourse in the media’ (quoted in Napoli and Gillis: 2008, 9).
This suggests that news and current affairs programming should, as a minimum, be considered but this does not exclude separate analysis of news and non-news content, and there are good grounds for more extensive review of this type. Indeed, focusing solely on news could lead firms to spin-off or sell their news services to avoid difficulties in the event of a planned merger, as envisaged in the proposed NewsCorp/Sky merger. Given the propensity for media operators to rely on a limited range of sources of news we also suggest that consideration should be given to wholesale, rather than simply retail, provision.
Although it may be easier to determine what is and is not ‘news’ or current affairs content than other programme genres, the process is unlikely to be uncontroversial. Three criteria have been suggested by the Dutch Media Authority: impact on the national population, a criterion it uses to exclude foreign outlets; generality of remit; and regular updating, at least once a week (Commissariaat Voor De Media: 2010, 88). The exclusion of foreign news sources that provide a different perspective on domestic foreign policy, can be criticised for taking too limited a view of what media pluralism entails (for discussion, see Arnott: 2010; Commissariaat Voor De Media: 2011, 86-87; Owen: 2004, 22). The German Media Concentration Authority, ‘KEK’, has suggested that foreign language television channels should be included in its market evaluation (Kek: 2007, 22). On the other hand, foreign news services are less likely to provide extensive coverage of domestic affairs and for this reason are more likely to be supplements rather than alternatives to domestic news services. ‘Generality of remit’ is also problematic in that it would exclude thematic news services such as a financial news channel that may have considerable social or political influence.
In the UK, Ofcom has developed relatively expansive definitions of news and current affairs programming for the purpose of establishing public service broadcasting quotas (House of Lords: 2012, 45). These include, alongside newscasts or bulletins providing national, international or regional news coverage, analysis of current events and issues, weather reports, programmes about topical business or financial issues, and coverage of parliamentary and political affairs.
In relation to geographic delimitation, specific markets should, as a minimum, be designated to reflect local, national (devolved) and UK-wide political constituencies. This is because the media play an important role in holding not only Westminster but also politicians at the devolved and local levels to account. National and local media services focus on different news content and cannot therefore be seen, in information terms, as substitutes.
d) The choice between fixed limits, thresholds and ‘holistic’ impact assessments
When formulating media ownership limits, the interests of both citizens and industry need to be taken into account. Although the interest of citizens in access to diverse and reliable information is here the pre-eminent regulatory concern, the ability to plan ahead and avoid undue delays remains important for industry. Regulatory costs also need to be kept within reasonable parameters. Where possible, and without compromising key objectives, therefore, ownership rules should be framed as clearly as possible and their mode of assessment readily understandable.
To achieve this, some sophistication in the rules may need to be sacrificed. The balance between predictability, speedy resolution and manageable costs, on the one hand, and regulatory sophistication, on the other, can be struck in at least three ways: through fixed ownership limits; through provisional thresholds that can be superseded in specific contexts; and through a wide-ranging plurality assessment along the line undertaken by Ofcom in its NewsCorp/Sky report. Fixed limits create the greatest certainty, while multi-factor investigations, though time consuming and costly, may result in more sophisticated and better targeted intervention.
Fixed limits offer not only a high level of predictability, they also limit regulatory discretion, reducing concerns over potential agency capture or political influence. They can be set at a level designed to guarantee protection of citizens’ interests. In the UK, Chris Goodall has suggested a return to fixed limits but rather than setting these by reference to the number of operators or audience shares, he proposes revenue-based controls, with no company allowed to obtain more than 15% of the revenues generated by the entire media sector (Goodall: 2012). This is similar to the Italian approach, where owners are prohibited from obtaining more than 20% of revenues derived from a broadly defined media sector, including audiovisual distribution services, book publishing and advertising agencies (Article 15 of Law 112/2004, discussed in Casarosa and Brogi: 2011, 22).
For the reasons elaborated in section 6(b) above, time-based metrics appear preferable to those based on revenue when addressing media power and influence, though a robust system for evaluating temporal exposure to radio and newspapers will need to be developed. This is in line with proposals put forward to Ofcom by the Daily Mail and General Trust (‘DMGT’), positing a simple 30% limit on share of news consumption (Sweney: 2012).
Fixed ownership limits, though providing considerable certainty for industry, cannot easily accommodate the full range of quantitative and qualitative factors that interact to enhance or reduce media plurality (Napoli and Gillis: 2008, 12-15). As a result, some countries have proposed converting fixed limits into thresholds that, when surpassed, trigger further consideration into whether the plurality concerns stemming from consolidation can be addressed by alternative means. A more exacting form of this approach allows the threshold to be exceeded only where positive benefits to the public interest can be shown.
This is illustrated by developments in the US, which has traditionally preferred ‘bright line’ rule-making and fixed limits in order to enhance certainty and reduce agency discretion (Goldfarb: 2004, n.34). In 2008, however, the FCC introduced new rules that designated certain combinations of print and television interests to be either presumptively consistent or inconsistent with the public interest. The rules allowed specified countervailing factors to be taken into account to rebut those presumptions. In relation to cross-media mergers between newspapers and television stations in the largest markets the rules established a presumption that media diversity would not be threatened where the television station was not ranked among the top four station and there would remain at least eight independent television, radio, newspaper or cable ‘voices’. For smaller markets there was a presumption against newspaper-television cross mergers.
This negative presumption could be ‘overcome’ by showing ‘clear and convincing evidence’ that the merger would increase the diversity of independent news outlets and increase competition among independent news sources. Moreover, the presumption would be automatically reversed in two circumstances, firstly, where the broadcast station involved in the merger had not previously offered local newscasts and undertook to provide seven hours a week of local news programming and, secondly, where the firm being taken over was no longer economically viable. In rebutting the various presumptions, four factors were specified to be relevant: the combined firms’ commitment to local news production; editorial independence and continuing commitment to separate staffing; the financial state of the firms and, where one was in distress, the scale of future investment in newsroom operations; and the level of concentration in the relevant markets. Although the rule changes were struck down in 2011 for lack of consultation in Prometheus II, the FCC has proposed the adoption of a similar case-by-case approach in the future (FCC: 2011).
Three features of the 2008 rules are particularly interesting. Firstly, they placed the burden on the merging firms, when a negative presumption was created, to make out a convincing positive case for further consolidation. Secondly, they clearly identified the relevant factors that would be taken into account. Thirdly, the negative presumption against consolidation was automatically reversed only in limited circumstances, namely, where there was a failing firm or where there was the offer of significant additional news services. If a similar approach had been applied to the News Corp/Sky case it is unlikely the consolidation would have been approved: Sky was not a failing firm and is one of the ‘top four’ television stations for news. Moreover, there was little indication that the merger would have positively enhanced media diversity, rather, attention focused on how the independence of BSkyB’s news service, and thus the status quo, could be preserved.
An approach along these lines for the UK has been proposed by the Co-ordinating Committee for Media Reform (‘CCMR’). CCMR proposes that cross-media mergers that would lead to, or exceed, a 15% share of news provision at national or regional levels, should be subject to a public interest review (CCMR: 2011, 17). For mergers exceeding the 15% threshold, the firms would be required to establish a countervailing public interest case for further consolidation. This is combined with a proposed 30% upper limit for each of the ‘radio, television, newspapers, online’ platforms that cannot be superseded. CCMR do not specify how the ‘share’ is to be computed and do not appear to envisage any form of weighting but, as suggested above, a share of news consumption metric could be employed.
Though a case-by-case approach of this type has certain attractions, important disadvantages remain. Firstly, as with fixed limits, if the threshold is based on an inappropriate or inadequate form of measurement the whole system will be flawed, regardless of the opportunity to make mitigating adjustments. Secondly, such approaches are not only less predictable and objective than those that employ fixed limits they also tend to advantage large media organisations with the resources to research and present a convincing case for exceeding the specified thresholds. Citizens or civil society organisations may find it difficult to obtain the information needed to rebut their position, particularly regarding prospective commitments. The discretionary element, affords greater scope for agency capture and for other interests, such as industrial concerns, to creep into the assessment. In practice, provisional thresholds may lead to a process of incremental consolidation, up to the fixed limits, that is difficult to reverse.
The third approach is to engage in a ‘holistic’, multi-factor analysis of market conditions, along the lines undertaken by Ofcom in its NewsCorp/Sky investigation. Though requiring further empirical and theoretical investigation, an exposure-based approach encompassing share and duration of reference, share of overall news consumption and multi-source data could provide a more sophisticated indicator of media power and potential influence. It would be possible to supplement these exposure-based metrics with more direct measures of influence from consumer surveys investigating recall, attitude modification and self-reported degrees of influence. The main problem with investigations of this type is that they are time-consuming, costly, and afford considerable discretion to the regulator. These problems could be ameliorated in part by co-ordinated collection, and regular publication, of relevant data as well as by the formulation of guidelines regarding the type and levels of consolidation likely to be considered problematic. In the NewsCorp/Sky merger, for example, Ofcom expressed concern not only over the scale of the combined firms share of reference but also the fact that the companies involved were among the four largest wholesale providers of news content, mirroring concerns in the US (Ofcom 2010b, para. 1.32).
Ultimately, therefore, the benefits of a multi-faceted plurality assessment in terms of targeted intervention may be outweighed by its complexity and attendant costs. Regular investigation on these lines could still, however, prove useful in monitoring media plurality in specific markets over time. The Dutch Media Authority, for example, has suggested that its opinion-power survey could be carried out on an annual basis (Commissariaat Voor De Media: 2011, 90). Ongoing independent review is also desirable to identify market concentration that arises other than through merger and acquisition activity, for instance, as a result of successful business strategies and economies of scale, which, from a citizen’s perspective, is equally problematic.
e) Fixing empirically and intuitively convincing thresholds
The development of appropriate forms of measurement should enhance the legitimacy of the review process but we still need to establish what degree of media concentration as a society we are prepared to tolerate, and this is as much a policy decision as an empirical question (Baker: 2009).
It is beyond the scope of this article to suggest specific thresholds and care should be taken when drawing conclusions from other countries given distinct market characteristics and policy preferences. Nevertheless, comparative research can offer valuable insights (Just: 2009). The FCC, for example, has commissioned a wide-ranging series of studies into the impact in the US of media ownership concentration on media diversity and this could form the basis for further empirical research and co-ordination at the European level. In particular, there is scope for regulators, industry and interest groups to explore best practice and research findings at the European level, supported by the European Union, which has shown increasing interest in media plurality issues.
In relation to the number of firms needed to guarantee sufficient diversity of information and prevent information being suppressed, Brocas et al hypothesise that at least four voices are required to represent opposing sides on any issue and to ensure that, in relation to each of these sides, alternative perspectives are put forward (Brocas, Carillo, Wilkie: 2011). But this was considered unlikely to be sufficient because there will often be more than two sides to a particular issue and not all outlets will address each topic. Their study concluded that there was indeed better transmission of information with four firms rather than two and that this improved further with six firms rather than four.
It is worth noting that the US requires more than six independent operators in certain contexts. In the 2007 rule revisions, for example, at least eight remaining independent voices were required for newspaper/television mergers to be acceptable in the largest media markets, while dual ownership of local television stations is only permitted if there remain eight independently owned stations (Renhoff: 2011, 2). But to require six independent (particularly news) voices in certain media sectors in the UK would be improbably ambitious given present market conditions. In the radio sector, for example, just two providers, Sky and the BBC, currently provide the vast majority of broadcast news, while consolidation in the regional press is well advanced given the difficult economic conditions (Ofcom 2010b: 37). These findings serve to underline the importance of considering other measures, including financial support and tax breaks, to enhance diversity in these consolidated sectors.
The US studies focus primarily on the impact of structural controls on media diversity but where concern is with undue media power, attention needs to focus on metrics that assess the relative capacity of firms to influence the public. We have suggested above that the time-spent accessing or being exposed to a particular source or programme type is the most convincing single metric currently available. As noted above, the CCMR has proposed a 15% threshold and a 30% limit in relation to share of news provision at national and regional levels; the DMGT has proposed a 30% exposure limit (Sweney: 2012); while Germany currently applies a 30% television audience share limit. The link between exposure and media influence clearly requires further investigation to enable the appropriateness of such real or posited limits to be assessed. A 30% limit would, however, allow just three owners to dominate access to news information, and from a diversity perspective, three owners falls short of the minimum four voices and well below the six voices identified as desirable in the Brocas study (Brocas, Carillo, Wilkie: 2011). If text, video and audio are qualitatively different media sources then separate cross and mono-media controls will also be required, to prevent complete consolidation in smaller media markets.
Public service providers should be included in any market assessment in order to prevent inappropriately inflating the shares, and thus curtailing the operations, of commercial firms (cf CCMR: 2011, 14). Publicly funded providers such as the BBC are deliberately insulated from commercial and political pressures that could lead to bias or the suppression of information and their commitment to ethical journalism and pluralistic policies go well beyond the basic statutory impartiality requirements and thus render scale less of an issue. Concerns over their impact on commercial operators can be addressed by more focused mechanisms, such as the public value test employed by the BBC Trust (BBC Trust: 2007), rather than through the application of general media ownership thresholds or limits.
In order to make the process of fixing thresholds more objective, a number of regulators and commentators have considered whether indices used to gauge market concentration levels can be adapted to assess media plurality (Noam: 2004; Hill: 2006; Morisi: 2012). The best know of these indices, the Hirschmann-Hirfindhal Index (HHI), uses as its measure the sum of the squares of the market shares of operators in a given market. The US Government considers an industry to be unconcentrated when the HHI is less than 1000, while markets with an HHI over 1,800 are defined as highly concentrated (Schwartz: 2010, 167). An attempt by the FCC in 2003 to adapt the HHI to assist in setting media ownership thresholds was subject to successful legal challenge in Prometheus I, largely because it treated all services within a particular media sector equally, regardless of audience share. In the light of this ruling various modifications have been proposed to the HHI, in particular, Noam has proposed enhancing the weight given to the number of voices in the market, alongside market share, by dividing the HHI by the square root of the number of voices (Hill: 2006, Noam: 2004). On this basis ‘the less concentrated and the more numerically diverse a market is’, the lower the index will be (Noam: 2004).
Application of the HHI to Ofcom’s 2010 share of reference figures in the NewsCorp/Sky case reveal that the UK media market was already highly concentrated before the merger (1,941 wholesale/1,821 retail), and this would have been further exacerbated if it had gone ahead. Similar conclusions can be derived from the application of other indices such as the Noam Index (Morisi: 2012). HHI thresholds have, however, been established with threats to market competition in mind and do not answer the prior policy question: when, from a plurality perspective, is concentration problematic? Although a high HHI score is suggestive that pluralism could be at stake, there may also be concentrations that are unproblematic from a competition perspective yet which still raise plurality concerns. The answer to this question requires a sound theoretical and empirical understanding of the point at which concentration starts to undermine diversity, increase bias or the suppression of information, or enables operators to exert an undue influence over public opinion. It would be possible to map the results of any such investigations back onto scales derived from relevant metrics, but the metrics alone cannot determine appropriate ownership limits or thresholds. And if we know what these limits are we do not need the indices to create additional certainty.
But indices of this type become more attractive in circumstances where, as now, empirical evidence is inconclusive or lacking. If Noam is correct that there is ‘no conceptual, practical or legal way to officially define and measure the vigour of a marketplace in ideas’, then an index linked to share of news consumption and number of independent sources could provide a clear and simple precautionary mechanism to address both media diversity and influence concerns (Noam: 2004). Use of such indices precludes the need to think through all potential configurations in advance and consequently reduces the complexity of the regulatory framework. Both the HHI and Noam indices also allow share and number of market participants to be configured in different ways while still falling within specific numerical bands, building a degree of flexibility into the rules. Application of these indices to share of news consumption figures, which we have suggested to be the most convincing single indicator of media influence, would also overcome the concerns expressed in Prometheus I that services with different audience shares were being treated equally. Alternatively, such metrics can provide a useful mechanism for monitoring variations in levels of consolidation over time.
f) Adequately resourced, specialist, independent review
UK procedures for evaluating media plurality not only involve multiple investigations by separate bodies, they also prioritise investigation by the generalist Competition Commission over that of the specialist media regulator Ofcom. Government also plays a decisive role in determining whether to initiate a media plurality investigation, its terms of reference, and whether a merger should ultimately be allowed to proceed. As recent allegations of government leaks regarding the NewsCorp/Sky investigation indicate, government involvement is likely to bring into question the integrity of the whole system.
There is much, therefore, to be said for vesting the power to initiate proceedings and apply the relevant rules in an independent, specialist media body, such as Ofcom. Council of Europe guidelines explicitly call for media regulatory bodies to be independent of both political and economic interests. The risk of undue industry influence can be reduced by allocating sufficient resources to enable the regulator to evaluate both market conditions and industry claims on an independent basis. Discretion can be limited by requiring that all concentrations exceeding the specified thresholds or triggers should be investigated, as was previously the case under the Fair Trading Act 1973. Similarly, factors that can mitigate plurality concerns should be set out as precisely as possible and the situations in which they can be taken into account clearly identified. Decisions by the regulator should be judicially reviewable.
Any legislator or regulator tasked with formulating media ownership controls currently faces two underlying problems: empirical uncertainty and causal complexity. To address the former and better understand the latter, additional research is clearly needed. Three issues, in particular, merit further investigation. Firstly, whether individuals do, in fact, relate differently to text, audio and video services; secondly, which factors can be used to measure influence; and, thirdly, the nature of the relationship between the number of available sources and viewpoint diversity. Answers to these questions would assist in determining whether we should continue to regulate, from a plurality perspective, different media markets separately; the appropriateness of specific forms of measurement, particularly those relating to exposure, discussed in section 6(b) above; and the minimum number of independent operators, if any, that should be required in any given sector. As noted, there is considerable scope for further collaborative work at the European level.
But pending answers that can be used to frame an empirically robust regime what steps should the regulator take? Our discussion above has highlighted the central importance of clarifying underlying objectives and, given the democratic and social issues at stake, the need for a precautionary approach. Where intervention seeks to maintain viewpoint diversity and prevent undue influence then both the number of independent sources and each operator’s share of exposure or other designated measurement of influence should be taken into account. We have suggested that the single most plausible measure currently available to assess potential influence is share of exposure to relevant sources. Consideration should be given, as a minimum, to news sources, broadly understood to include current affairs and documentary programming, but also to programme services from particular sources more generally. Concentration in the wholesale provision of news should also be assessed along Ofcom lines.
Such a measure could be strengthened by additional information regarding the regularity and duration of each engagement and the degree of multi-sourcing. More direct indicators of influence, such as recall, attitude modification, even personal assessments, are also potentially relevant, though may not be sufficiently reliable to act as a basis for intervention. There is, moreover, a real risk, given our current state of knowledge, that in enhancing the ‘sophistication’ of our regulatory regime we render it not only more open to legal challenge but also, in practical terms, unworkable.
Fixed limits that can be easily applied using readily available data are thus inherently attractive in that they reduce regulatory discretion and enhance certainty for industry. For this reason we suggest that only in exceptional cases, notably where a failing firm is involved, should there be scope to supersede the specified limits, and only on the basis of meaningful undertakings designed to enhance the range and quality of available content. In order to operate effectively, reliable measures of news exposure for the printed press and radio will need to be developed. In principle, such limits are equally relevant whether or not growth stems from merger and acquisition activity or a firm’s successful business strategy, so that regular market review will be required. Indices derived from the field of competition law, such as the Noam Index, applied to share of news consumption figures, could provide a relatively clear, simple and flexible mechanism for monitoring concentration trends.
Presumptive thresholds, along the lines explored by the FCC in the US and the CCMR in the UK, though allowing a wider range of considerations to be taken into account, could in practice lead to undesirable levels of consolidation with any upper limit becoming the de facto point of control. Rather than extracting concessions solely from firms seeking to consolidate, measures to promote quality journalism should be pursued across the board. Concerns over the ability of powerful media owners to influence government policy-making, for example, can be addressed, at least in part, through the adoption, and effective enforcement, of appropriate professional standards for journalists and politicians, as well as transparency in the lobbying process. Similarly, support for a mixed media environment, with diverse operators comprising non-profit and charitable organisations, as well as publicly financed public service institutions or services, can help to counter potential bias and enhance viewpoint diversity.
Given our present state of knowledge, the framing of media ownership limits undoubtedly remains more of an art than a science. Moreover, the complexity of causal relations in this field means that controls of this type are only ever likely to be one among a wide range of strategies employed to enhance the range, diversity and quality of media content. But when incorporated into a broader regulatory framework of this kind, media ownership limits can play their own distinctive part in strengthening the democratic process and enhancing individual freedom and autonomy.
Alexander, P.J. and Brown, K (2007), ‘Policymaking and Policy trade-offs: Broadcast Media Regulation in the United States’ in Seabright and Von Hagen (2007) 255-279
Arnott, C. (2010), ‘Media Mergers and the Meaning of Sufficient Plurality: A Tale of Two Acts’, (2010) 2 The Journal of Media Law 245-276
Baker, C.E. (2007), Media Concentration and Democracy, Why Ownership Matters (Cambridge University Press, 2007)
Baker, C.E. (2009), ‘Viewpoint Diversity and Media Ownership’ (2009) 60/3 Federal Communications Commission Law Journal 651-672
Ball, J. (2011), The Conservatives’ contact with News International, The Guardian ‘Datablog’, Wednesday 27 July 2011 at:
Barendt, E. (2005), Freedom of Speech (Oxford University Press 2005 second edition)
Beebe, J. (1977), ‘Institutional Structure and Program Choice in Television Markets’ (1977) 91 Quarterly Journal of Economics 15
Bernhardt, D., Krasa, S., and Polborn, M. (2008), ‘Political Polarization and the Electoral Effects of Media Bias’ (2008) 92 Journal of Public Economics 1092-1104
British Broadcasting Corporation Trust (2007), Public Value Test (PVT): Guidance on the Conduct of the PVT, August 2007, at: http://www.bbc.co.uk/bbctrust/assets/files/pdf/regulatory_framework/pvt/pvt_guidance.pdf
Brocas, I., Carillo, J.D., Wilkie, S. (2011), ‘A Theoretical Analysis of the Impact of Local Market Structure on the Range of Viewpoints Supplied’, FCC Media Study no. 9, June 2011, at: http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-307525A1.pdf
Buncombe, A. (2007), ‘US military tells Jack Bauer: Cut out the torture scenes … or else!’ The Independent on Sunday, Tuesday, 13 February 2007
Casarosa, F. and Brogi, E. (2011), ‘Does Media Policy Promote Media Freedom and Independence?’, Case Study Report for the Mediadem project, December 2011 at: http://www.mediadem.eliamep.gr/uploads/2012/01/Italy.pdf
Castells, M. (2010), ‘Communication Power: mass Communication, Mass Self-Communication, and Power Relationships in the Network Society’ in Curran (2010) 3-17
Commissariaat Voor De Media (2011), Mediamonitor. The Dutch Media in 2010, February 2011, Commissariaat Voor De Media, Hilversum, The Netherlands at: http://www.mediamonitor.nl/dsresource?objectid=11689&type=org
Competition Commission (2007), Report on the Acquisition by British Sky Broadcasting Group PLC of 17.9% of the Shares in ITV plc, 14 December 2007, available at:
Congdon, T., Graham, A., Green, D., Robinson, B. (1995), The Cross Media Revolution, Ownership and Control (John Libbey, London, 1995)
Co-ordinating Committee for Media Reform (2011), ‘The Media and the Public Interest’, Briefing Paper, 4 November 2011, available at:
Cooper, M. (ed.)(2007), The Case Against Media Consolidation: Evidence on Concentration, Localism and Diversity, Donald McGannon Center for Communications Research, Fordham University, available at:
Cooper, M. and Cooper, S. (2007), ‘Concerns about Print Journalism and Cross-Ownership’, Study 10, in Cooper, M. (ed.)(2007) 159-180
Cooper, M. (2007), ‘Local Media and the Failure of the Consolidation/Conglomeration Model’, Study 9, in Cooper, M. (ed.)(2007) 137-158
Craufurd Smith, R. (2009), ‘Media Ownership and the Public Interest: the Case of Virgin Media, British Sky Broadcasting and its ITV Shares’ (2009) 1 The Journal of Media Law 21-36
Craufurd Smith, R. (2004), ‘Rethinking European Union Competence in the Field of Media Ownership: the Internal Market, Fundamental Rights and European Citizenship’, (2004) 5 European Law Review 652-672
Curran, J. (2010)(ed), Media and Society (Bloomsbury Academic, London and New York, fifth edition)
Czepek, A. and Klinger, U. (2010), ‘Media Pluralism Between Market Mechanisms and Control: The German Divide’ (2010) 4 International Journal of Communication 820–843
DellaVigna, S. and Kaplan, E. (2007) ‘The Fox News Effect: Media Bias and Voting’ (2007) 122/3 Quarterly Journal of Economics 1187-1234
Department for Culture Media and Sport (2001), Consultation on Media Ownership Rules, November 2001, DCMS, London
Department of National Heritage (1995), Media Ownership: The Government’s Proposals, Cm 2872, HMSO, London
Department of Trade and Industry (DTI) (2004), Guidance on the operation of the public interest merger provisions relating to newspaper and other media mergers, May 2004 at: http://www.bis.gov.uk/files/file14331.pdf
Dyke, G. (2004) Inside Story (Harper Collins, London 2004)
Enders, C. (2010), ‘News Corporation’s Proposed Takeover of BSkyB: A Submission to the Secretary of State by Claire Enders, CEO, Enders Analysis Ltd’, 30 July 2010, available at: http://image.guardian.co.uk/sys-files/Media/documents/2011/08/24/foi-bskyb-7-enders-letter-31-july-2010.pdf.
Erb, J. (2011), ‘Local Information Programming and the Structure of Television Markets’, FCC Media Ownership Study 4, May 2011, at:
Federal Communications Commission (2011), Notice of Proposed Rulemaking, 22 December 2011, FCC 11-186
Freedman, D. (2008), The Politics of Media Policy (Polity, Cambridge 2008)
Freedland, J. (2011), ‘The New Met Chief’s U-turn is Welcome – He Had Made a Gross Misjudgment’, The Guardian, 21 September 2011,
Gentzkow, M. and Shapiro, J.M. (2006), ‘Media Bias and Reputation’ (2006) 114 Journal of Political Economy 280-316
George, L.M. and Oberholzer-Gee, F. (2011), ‘Diversity in Local Television News’, FCC Media Ownership Study 8b, 27 May 2011, revised 13 July 2011 at: http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-308602A1.pdf
Goldfarb, C.B. (2004), FCC Media Ownership Rules: Current Status and Issues for Congress, CRS Report for Congress, August 2004, 1-38
Goodall, C. (2012), Response to Ofcom’s Invitation to Comment on Measuring Media Plurality, 5 January 2012, at:
Grice, A. (2011), ‘Blair says he wasn’t too close to Murdoch. I disagree’, 28 July 2011, Independent.co.uk
Hill, B.C. (2006 ) ‘Measuring Media Market Diversity: Concentration, Importance, and Pluralism’, 58 Federal Communications Law Journal 169-194
Hindman, M., (2011), ‘The Lack of Local News on the Internet’, research paper prepared for the FCC, FCC PUR11000027, available at: http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-307524A1.pdf
Hitchens, L. (2006), Broadcasting, Pluralism and Diversity, A Comparative Study of Policy and Regulation (Hart Publishing, Oxford, Portland, 2006)
Ho, D. and Quinn, K. (2009), ‘Viewpoint Diversity and Media Consolidation: An Empirical Study’ (2009) 61 Sta. L. Rev. 781
House of Lords, (2012), The Future of Investigative Journalism, Third Report of the Select Committee on Communications, Session 2010-12, HL 256
Hunt, J. (2011a), ‘A communications review for the Digital Age’ (open letter), 16 may 2011, available at:
Hunt, J. (2011b), ‘Boldness Be My Friend’, speech to the Royal Television Society, Wednesday 14 September 2011 at:
Iosifidis, P. (2010), ‘Pluralism and Concentration of Media ownership: Measurement Issues’ 17 javnost-the public, 5-22
Iyengar, S. (2010), ‘The State of Media Effects research’ in Curran, J. (2010)(ed) 273-289
Just, N. (2009), ‘Measuring Media Concentration and Diversity: New Approaches and Instruments in Europe and the United States’, 31 Media, Culture and Society, 97-117
Joint Committee (House of Lords and Commons) on Draft Communications Bill (2002), Report, HL 169-1/HC 876-1, vol. 1
Karppinen, K. (2010), University of Helsinki, Department of Social Research, available at:
KEK (2007), 3rd Report on Media Concentration ‘Cross-Media Relations:
A Challenge for Media Concentration Control’ (3. Konzentrationsbericht der KEK: “Crossmediale Verflechtungen als Herausforderung für die Konzentrationskontrolle”), 2007, Schriftenreihe der Landesmedienanstalten, Band 36, VISTAS Verlag Berlin, available at: http://www.die-medienanstalten.de/profile/organisation/the-commission-on-concentration-in-the-media-kek.html
Kelley, D. and Donway, R. (1990), ‘Liberalism and Free Speech’ in Lichtenberg, J (ed.) (1990), chapter 2
Lichtenberg, J (ed.) (1990), Democracy and the Mass Media (Cambridge University Press, Cambridge, 1990)
Lichtenberg, J. (1990), ‘Foundations and Limits of Freedom of the Press’ in Lichtenberg, J (ed.) (1990), chapter 3
McQuail, D. (2010), Mass Communication Theory (Sage, Los Angeles, London, 2010)
Morisi, D. (2012), Response to Ofcom consultation on measuring media plurality, 5 January 2012, at:
Mulholland, H. (2011), ‘Cameron refuses to deny discussing BSkyB bid with News International’, The Guardian, Wednesday 20 July 2011 at guardian.co.uk
Napoli, P. and Gillis, N. (2008), ‘Media Ownership and the Diversity Index:
Outlining a Social Science Research Agenda’, McGannon Center Working Paper Series, Paper 5
Napoli, P. (2011), Audience Evolution. New Technologies and the Transformation of Media Audiences (Columbia University Press, New York, 2011)
News Corporation (2011), Submission to the Secretary of State for Culture, Olympics, Media and Sport, 14 January 2011, at:
Noam, E. (2004), ‘How To Measure Media Concentration’, FT.COM, 30 August 2004 at: http://www.ft.com/cms/s/da30bf5e-fa9d-11d8-9a71-00000e2511c8.html
Ofcom (2007), Report for the Secretary of State pursuant to Section 44A of the Enterprise Act 2002 of British Sky Broadcasting plc’s acquisition of 17.9% shareholding in ITV plc, 27 April 2007 http://www.bis.gov.uk/files/file39607.pdf
Ofcom (2010), Pay TV Statement, 31 October 2010 at:
Ofcom (2010b), Report on public interest test on the proposed acquisition of British Sky Broadcasting Group plc by News Corporation, 31 December 2010, available at: http://stakeholders.ofcom.org.uk/binaries/consultations/public-interest-test-nov2010/statement/public-interest-test-report.pdf, 6-7
Ofcom (2010c), Response to Secretary of State: Local Media – Cross Media Ownership Rules, 9 August 2010
Owen, B.M. (2004), ‘Confusing Success with Access; “Correctly” Measuring Media and Online Concentration’, Stanford Law and Economics Olin Working Paper no. 283, May 2004
PEW (2010), The State of the News Media, An Annual Report on American Journalism at: http://stateofthemedia.org/2010/online-summary-essay/
Polo, M. (2007), ‘Regulation for Pluralism in Media Markets’ in Seabright and Von Hagen (2007), 150-188
Renhoff, A.D. and Wilbur, K.C. (2011), ‘Local Media Ownership and Viewpoint Diversity in Local Television News’, FCC Study on Media Ownership 8A, June 2011 at: http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-308596A1.pdf
Robinson, B. (1995), ‘Market Share as a Measure of Media Concentration’, in Congdon et al (1995), 50-72
Royal Commission on the Press (1949), Report, CMD 7700, (HMSO, London, 1949)
Seabright, P. and Von Hagen, J (2007)(ed), The Economic regulation of Broadcasting Markets, Evolving Technology and Challenges for Policy (Cambridge University Press, Cambridge)
Scheuer, A. (2006), ‘Media Law Reservations about Springer/ProSiebenSat.1 Merger’, IRIS 2006-2:9/13
Smith, J. (2009), ‘Monsanto Forced Fox TV to Censor Coverage of Dangerous Milk Drug’ The Huffington Post, 13 April 2009
Steiner, P. (1952), ‘Program Patterns and the Workability of Competition in Radio Broadcasting’, (1952) LXVI Quarterly Journal of Economics 194-223
Schwartz, E.C. (2010), ‘Conflicts of Interest: How Media Pluralism Protects Democracy and Human Rights’, 2010 at: http://gradworks.umi.com/3423027.pdf
Sunstein, C. (2000), ‘Television and the Public Interest’ (2000) 88 California Law Review 501
Sweney, M. (2012), ‘Daily Mail Publisher: Cap Media Groups’ Share of News Provision’, 5 January 2012, the Guardian at: guardian.co.uk
Turner, T., (2004), ‘My Beef with Big Media’ (2004) Washington Monthly, July/August
Valcke, P., Picard, R., Sükösd, M., Klimkiewicz, B., Petkovic, B., Dal Zotto, C., Kerremans, R. (2010), ‘The European Media Pluralism Monitor: Bridging Law, Economic and Media Studies as a First Step Towards Risk-Based Regulation in Media Markets’, (2010) 2 The Journal of Media Law 85-113
Waldfogel, J. (2002), ‘Consumer Substitution Among The Media’, FCC Media Ownership Working Group paper, September 2002, at: http://transition.fcc.gov/ownership/materials/already-released/consumer090002.pdf
Waldfogel, J. (2011), ‘Station Ownership and the Provision and Consumption of Radio News’, FCC Media Ownership Study 5, 6 June 2011 at: http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-307470A1.pdf
Westphal. D. (2002), ‘Media Pluralism and European Regulation’, (2002) European Business Law Review 459
Yliniva-Hoffmann, A. (2012), ‘BayVGH Upholds Axel Springer AG Complaint about P7S1 Takeover’ IRIS 2012-4:1/15
* University of Edinburgh, Script Centre, and London School of Economics respectively. This paper was presented at a conference organised by the Centre for Media Pluralism and Media Freedom at the European University Institute, Florence, on 27-28 April 2012 and a short version will be published in the first issue of The Journal of Media Law (2012). We would like to thank the Centre for creating a venue for discussing these ideas as well as the support provided by the European Union for the Mediadem project, examining policies to support free and independent media, in which the University of Edinburgh is a participant. In particular, we wish to thank Davide Morisi for his helpful research in relation to the various indices discussed in this paper.
 App No. 13914/88; 15041/89; 15717/89; 15779/89; 17207/90, Informationsverein Lentia and Others v. Austria, (1993) 17 EHRR 93, and, for a more accommodating approach, see Tele 1 Privatfernsehgesellschaft v Austria (2002) 34 EHRR 181.
 Key documents can be located at: http://webarchive.nationalarchives.gov.uk/20101227023510/http:/www.bis.gov.uk/policies/business-law/competition-matters/mergers/mergers-with-a-public-interest/broadcasting-and-cross-media-mergers. See also British Sky Broadcasting Group Plc, Virgin Media Inc v The Competition Commission, The Secretary of State for Business Enterprise and Regulatory Reform  EWCA Civ 2; 2010 2 All E.R. 90.
 Key documents and details regarding the development of the investigation can be obtained at the DCMS website at: http://www.culture.gov.uk/news/news_stories/7738.aspx.
Rupert Murdoch, for instance, has significant interests in the field of book publishing, notably Harper Collins, which has supported authors in line with his own political leanings, though it has also published books critical of his media influence (McKnight and Hobbs: 2011; Dyke: 2004).
 British Sky Broadcasting Group Plc, Virgin Media Inc v The Competition Commission, The Secretary of State for Business Enterprise and Regulatory Reform  EWCA Civ 2; 2010 2 All E.R. 907, para.90.
 See Valcke et al: 2010, and details of the Centre for Media Pluralism and Media Freedom and the High Level Group on Media Freedom and Pluralism at: http://ec.europa.eu/information_society/media_taskforce/pluralism/index_en.htm.