If Mark Twain were offering advice today, he’d probably say ‘buy time, they’re not making any more of it.’ Certainly the major internet intermediaries are successfully corralling our time, intent on keeping us with them so that they can monetise their consumption metrics for advertisers and shareholders. As Tim Wu points out in his compulsive book The Attention Merchants, which takes us into the heart of behavioural advertising, ‘in this game of trackers and profile builders, as in so many others, Google and Facebook, de facto diarchs of the online attention merchants, reigned supreme.’
A study by Enders Analysis last month found that Facebook and Google accounted for 90% of the growth in UK digital advertising in 2016. Earlier studies have shown the duopoly accounting for up to 85% of US digital advertising revenues. Even though Facebook recently found itself under pressure after admitting it had miscalculated its metrics, it seems unlikely advertisers are going to give up on these platforms. The ad revenue of old media has declined in direct proportion to the growth of ad revenue for Google and Facebook. Facebook’s growth of course follows the introduction of Smartphones in 2007, as John Naughton has noted.
As well as their dominance of advertising, the two ‘titans’, as Professor Diane Coyle calls them, have become the dominant news distributors as well. 44% of US adults get their news via Facebook according to the Pew Research Centre having taken over as the top news referrer from Google in 2015 according to the traffic analytics site parse.ly. At least originating news organisations get to keep their branding in the Google News app: in the Facebook News Feed, as Alex Hern pointed out in the Guardian, there’s no branding difference between fake news sites and established and respected news outlets:
Facebook pulls the top image, headline, and normally an introductory paragraph, and formats it in its own style: the calming blue text, the standard system font, and the picture cropped down to a standard aspect ratio.
Nor, of course, is there any branding in Google searches, meaning that fake news can vie with real news for top spots.
So, can anything be done? Yes, and that leads me to the legislative imperative. I approach this with some hesitancy, having not had even the most minor involvement in media policy-making for almost a decade, since Rhodri Morgan made me a Minister in the Welsh Government back in 2007, and with considerable humility, recognising that many other people have been advocating a regulatory agenda in academia and outside to promote media plurality. I’m particularly thinking of work by Damian Tambini and Sharif Labo, Des Freedman , Martin Moore , Robin Mansell, and Justin Schlosberg , as well as my former BBC colleague, Robin Foster and many others. Since the US election threw a new focus on ‘fake news’ on social media, I’ve been reading myself back into the field, and I’ve learned a lot from a variety of academic sources in the UK and the rest of Europe, and in the United States. I’m particularly grateful to my former BBC European Affairs colleague Patricia Galvin, who subsequently worked at NTL, Ofcom and RTE, for her in-depth telephone tutorial on the current trajectory of EU media policy-making, and for her advice on this article. I’m also grateful to Professor Richard Sambrook of Cardiff University’s school of Journalism, Media and Cultural Studies (JOMEC) for inviting me to sit in on a fascinating presentation by Nic Newman of the Reuters Institute for the Study of Journalism on the Digital News Project 2016.
Sometimes the talking has to stop. As I read back through debates on media policy over the last few years it is obvious that opinion is divided between the internet intermediaries themselves — who continue to argue that they are not media companies, are widening the base of user-generated content and providing platforms for media companies to reach new audiences, and are promoting competition, even as they amass the bulk of growth in the digital media advertising markets — and a wide variety of other media organisations, civil society activists and academics who clearly believe that media pluralism is under threat, that there are new issues of power, concentration and dominance not adequately captured in existing competition rules or tests, and that action is needed. The immediate forum for that action, ironically given 2016’s events, is probably the European Union and its Digital Single Market agenda.
What seems missing, however, as far as I can tell from my reading and conversations, is either the strategic industry alliance necessary to drive forward an agenda to address the powers of internet intermediaries, in terms of content rules, competition issues and their dominance of the advertising markets, which as we have seen has had the effect of undermining the newspaper industry in particular — or any consensus on key planks of any legislative initiatives. It may be that industry players are more concerned with the deals that they can strike with the dominant internet intermediaries; it may be that there is nervousness about a legislative approach which could be counter-productive, such as the Spanish Google News example; it may be that jurisdictional complexities render any European solution difficult; and meanwhile we are in the latter days of some elements of the current EU legislative process. I well understand that with emerging industries, choking off the flow of investment is undesirable. But we’re past the stage of needing to secure a flow of early investment for Internet Intermediaries, which, as Coyle has said in the case of Google and Facebook ‘go far beyond any other commercial entities in the scale and dominance they have achieved ‘.
I recall twenty years ago, when the BBC and other free-to-air broadcasters were concerned about the emerging digital platforms and their potential controls of access to new digital services, the industry didn’t stay silent. I was the BBC’s head of public affairs at the time: along with others, we set in train a campaign on conditional access systems which resulted in the European Parliament, with its new co-decision powers under the Maastricht Treaty, and using the new Maastricht-granted responsibility for the EU to intervene in the field of culture, achieving amendments to the Television Standards Directive which passed into law despite being opposed by both the UK Government and the EU Commission. These granted all broadcasters access on ‘a fair, reasonable and non-discriminatory basis’ to TV decoders and platforms. Subsequent campaigning delivered amendments to the recitals of the 1997 TV without Frontiers Directive in respect of public and private free-to-air channels’ access to decoders and also recognition of the need to protect public access to major sporting events. At the same time there was a successful campaign to drive forward protections for public service broadcasters in the Treaty of Amsterdam, but that’s a longer article for another day.
Now, it is certainly more complicated lobbying in an EU of 28 member states, than one of twelve or fifteen. But I worry, as I read my way back into this area, that the increasing emphasis on self-regulation in media and technology policy over the last decade has served to de-politicise and make issues of media power and concentration less visible — Leveson aside — and accessible, in a regulatory or legislative sense, only to those with significant capital in terms of knowledge of technological capability or of competition law. While regulators such as Ofcom have made real attempts to gather the views of the public, the nature of the regulatory process can be quite a closed one or indeed such authorities may be restricted by their own scope and remit. The remedy to that of course is re-politicisation — the definition of a short and clear agenda which can be pursued through whatever legislative routes are immediately open. Sometimes that will be, to use a currently fashionable term, messy. Regulators don’t like mess, but politicians recognise, as President Obama said recently, that ‘democracy is messy’. Legislation is a process, not an event. Sometimes the process of making legislation, messy as it is, is as important as the passage of it. Indeed, the threat of legislation can sometimes change the behaviours of dominant players.
So let me make a few suggestions. First, I wholly support the idea that at a national level there needs to be remedial action, such as a levy on internet intermediaries’ revenue (effectively on advertising income) to support plurality in media, as suggested by the Media Reform Coalition. This underlines the case for endorsing EU Commission proposals to empower national regulators to impose levies on online platforms targeting their countries.
Second, we need some form of confidential independent regulatory scrutiny of proprietary data, including that held by Internet Intermediaries, as suggested by Tambini and Labo: this is as important to the interests of advertisers as it is to news outlets themselves: there is industry cooperation to collect TV and radio data, so why not internet data? (There is also a much wider debate around the public nature of data as infrastructure as Evgeny Morozov has argued recently). Philip Howard says that data in the hands of Facebook and Twitter is needed publicly to help make democracy thrive.
Third, we need to begin a more serious debate on algorithmic accountability, as recent work by Cadwalladr, Napoli and Satell have suggested. The draft Audiovisual Media Services Directive does refer to algorithms in the context of video-sharing platforms and their responsibilities in relation to hate speech and the protection of minors, opening the door for further scrutiny by national regulators. Under the Digital Single Market programme, work has started on algorithms in the context of online platforms. This is referenced in the Commission staff working paper on Online Platforms. The Commission has started work on this and there was a recent meeting in the European Parliament on the issue. Separately from the EU, the Council of Europe has established a Committee of Experts on Internet Intermediaries whose terms of reference include the examination of algorithmic accountability. Indeed, Ben Wagner, Rapporteur for the Committee of Experts, points out that there are already forms of regulation of algorithms in certain situations — and algorithmic regulation is likely to throw up even bigger challenges in the future.
My fourth proposal would be legislative action to consider imposing ‘due prominence’ — a concept developed in respect of electronic programme guides on digital television — for material produced by genuine originating news organisations in Facebook’s News Feed and similar systems, which would ensure as a minimum that the branding of the originating news outlet is incorporated within the Feed rather than the situation, as at present, where Facebook re-brands everything to fit its own standard. I would be more sceptical of copyright actions unless the problems of the Spanish Google news arrangements can be averted.
Finally, there needs to be effective recognition that certain Internet Intermediaries or Online Platforms are media organisations — in Ofcom’s words, editorially responsible providers — and should at least carry responsibilities in respect of protection of minors and prevention of hate speech. That requires an extension of the scope of the revised AVMS Directive and a clear definition of specific Internet Intermediaries, using the principle of graduated regulation, not to suggest that they should be covered by every aspect of the Directive. This will of course be resisted by many of those players who are clear that they are not media companies: I will write more about that in due course, but the German Justice Minister has already called for Facebook to be treated as a media company. It may be that the legislative route would be to revise the e-Commerce Directive, which, as the European Commission noted in its Communication on Online Platforms ‘was designed at a time when online platforms did not have the characteristics and scale they have today’.
The House of Lords European Union Committee’s Report on Online Platforms and the Digital Single Market warned of political pressure building over the future of online platforms and suggested that the Commission’s consultation was too broadly designed to address such questions decisively (paras 387–395) and raised the question of ‘whether enforcement agencies have the necessary powers and resources to act against abuse by the largest online platforms’. The Lords were right. This is not an issue that will go away, and nor should it.
Others with more legal and technological expertise may wish to question some of this, which I regard as a work in progress, and indeed may have other solutions. Moving forward, there needs to be a coordinated and sustainable lobby at a European level, involving media organisations, advertisers, civic society organisations, and academic specialists interested in media policy. I honestly believe that there is significant potential for an alliance of citizens and civic society organisations along with established media groups and advertisers and advertising agencies to create the space for legislative action
· In defence of facts on digital advertising metrics
· In defence of facts in news reporting and/or attribution
· In defence of the rule of law (for example German hate speech laws)
In practice, the nature of proposals will change and vary according to the internal debates and legislative approaches available. But the pending EU Single Market legislation, in respect of the AVMS Directive , and the Communication on Online Platforms and the Digital Single Market Opportunities and Challenges for Europe, provide a basis for action in the European Parliament. And I note even the EPP group has concerns about the role of Internet Intermediaries:
The EPP Group is strongly concerned that the growth of traditional media on the digital market is challenged by some news aggregators and search engines that develop their activities by using right-holders’ content without contributing to its development and without ensuring fair remuneration of the creators. The EPP Group is in favour of a clarification of the legal status, the role and responsibility of these platforms and content providers.
There is a final point here, about post-Brexit law-making. Assuming Brexit goes ahead, and the UK does want a relationship akin to the EEA, then it’s likely it will have to adhere in practice to EU Media laws. EU legislation may be our last, best hope for effective action. There’s a thing.
Leighton Andrews is professor of practice in public service leadership and innovation at Cardiff Business School; a former Welsh Government Minister, and a former head of public affairs at the BBC.