Online Political Communication


The Commission will consider what a ‘good’ political campaign would look like in the platform era, and how platforms, the media, government and civil society might work together to achieve it.


*A workshop with practitioners was held on 27th April 2018, you can read the summary report here*

‘1998: The Internet will connect us and allow us to share all kinds of information with those we care about!

2018: I answered a personality quiz and my friends' personal info was weaponized by a shadowy organization in order to try and rig an election.’

(Mathew Ingram (@mathewi) 11/04/2018)

Early internet thinkers were sometimes evangelical about the ability of online networks to democratise debate, and especially to connect activists (Clay Shirky, Here Comes Everybody, 2008). In the UK, MySociety’s was perhaps the best-known attempt to open up the workings of Parliament to scrutiny, and the Freedom of Information Act 2000 was similarly optimistic about the effectiveness of public scrutiny. But in recent years, concerns have emerged about the impact social media is having on political deliberation. The EU referendum and the US presidential race in 2016 both featured misinformation that spread rapidly through social networks, and was further disseminated by mainstream media. In the US, Russian actors were able to amplify these trends by using bots and memes. The Commission’s focus, however, is on the structural trends that allow foreign disinformation to spread, rather than Russian interference per se.

More widely, worries about the polarising effects of ‘filter bubbles’ and ‘echo chambers’ perpetuated by social media grew (Cass Sunstein, Republic 2.0 and #Republic), though some researchers play down their significance, finding that social media users see news from more sources than do consumers of traditional media. In Britain, these concerns were accompanied by growing dissatisfaction with the state of democratic deliberation and voting systems. First-past-the-post, ‘winner takes all’ voting systems, direct democracy and the impartiality of the BBC came under attack from all sides of the political spectrum. ‘Online engagement’ became synonymous with the low-effort gestures of likes, shares and e-petitions. Populist policies and governments have made significant inroads in Europe and the US.

In March 2018 the Observer reported that consultancy firm Cambridge Analytica had been working with a Cambridge University academic to ‘harvest’ Facebook data from at least 87 million people, and use it to target political advertising in the 2016 presidential campaign. The users affected had taken a personality quiz, or were friends with someone who had. The discovery prompted urgent questions about the ability of Facebook to use psychometric data to target political advertising, at relatively low cost to campaigners:


  • How many data points does Facebook hold about its users, and how did advertisers exploit these for political rather than solely commercial purposes ‘to target their inner demons’, as a whistleblower told the Observer?
  • To what extent are voters aware of how they are being targeted? (The WhoTargetsMe Facebook tool enables people to do this, but is little known among the wider public.)
  • What effect does microtargeting have on our ability to test political claims during a campaign?
  • If it does increase political engagement, is that at the cost of informed deliberation, creating an increasingly polarised political culture?
  • Does ‘building community’ in practice mean an atomised politics, often conducted in secret, and polarisation?


Facebook responded to the Cambridge Analytica revelations by making changes to the way it runs political advertising:


  • Advertisers will be verified by post before they can post
  • Users will be able to see who paid for a political ad, and their affiliation
  • These regulations will be extended to ‘issue’ groups (such as those campaigning in the Irish referendum on abortion)

The company is also working with several foundations to make its data more available to researchers.

In interviews and statements after the Cambridge Analytica revelations, Mark Zuckerberg indicated Facebook would work more closely with governments to tackle the spread of misinformation. Noting that the network had itself taken on some of the characteristics of a government - with the need for internally consistent rules that this implies - he acknowledged the difficulty in producing multinational policies: ‘How can you set up a more democratic or community-oriented process that reflects the values of people around the world?’ Zuckerberg has also spoken of the need to foster civic engagement at a local level through supporting news outlets.

Moves by governments

In the UK, the Electoral Commission is investigating whether Vote Leave effectively exceeded spending limits during the referendum by channelling funds to non-official campaigners that then spent heavily on Facebook. Electoral campaigning laws drafted in the pre-social media era have arguably proved unfit for digital campaigning (see Damian Tambini et al, The new political campaigning). For example, Rules for Party Election Broadcasts, look increasingly bizarre when campaigners can rapidly accumulate millions of views for a viral video. In France, for example, there are proposals to shut down social media networks if the spread of misinformation is thought to threaten the integrity of an election campaign.

But there is resistance to giving governments more powers over social media, particularly at a time when many media outlets are regarded as biased and parts of the EU (notably Hungary) are vulnerable to authoritarian backsliding. For the same reason, Facebook’s plans to work more closely with governments are potentially problematic, particularly if co-operation is done in an effort to pre-empt regulation and avoid democratic scrutiny.

Others ask whether government intervention - perhaps by backing alternative social networks, or at least breaking up existing ones - could produce healthier spaces for political deliberation. Still other commentators have blamed a complacent political establishment for ignoring large numbers of voters and thereby stoking cynicism and an openness to conspiracy theory (the so-called ‘redpill’ effect, in which users of online forums like Reddit try to ‘disabuse’ new users of their assumptions about who controls public discourse). Tweaks to electoral law, they argue, are inadequate to tackle a systemic crisis in public trust in politics.

Challenges to traditional media coverage

At the same time, the BBC has come under particular attack for its coverage of the Scottish referendum, Brexit and the Labour Party under Jeremy Corbyn. Faced with a proliferation of sources and perceived bias from the media, some of the public say they want to get information directly from politicians - a realisation that has served Donald Trump well on Twitter.

This preference for disintermediation fundamentally threatens the ability of media to report on campaigns, either because it has a chilling effect, or because viewers and readers abandon outlets that invest money in covering elections. Familiar formats such as the town hall debate, vox pops, opponents going head-to-head and (more recently) leaders’ debates have the potential to move opinion and shape debate, but exist in a hybrid media environment where voters are also getting information and engaging in discussion on social media.

The question of whether the media has a duty to fact-check claims made during election campaigning is acute, particularly since the Advertising Standards Authority does not hold political advertisers to account. This debate extends to the duty of the platforms. Where once they were keen to stand above the political fray and avoid taking responsibility for the content of posts, they now face pressure to act. Yet Facebook’s experiment with flagging disputed stories prompted resistance from users, and was abandoned in favour of a ‘Related Stories’ tool.

The Commission will consider what a ‘good’ political campaign would look like in the platform era, and how platforms, the media, government and civil society might work together to achieve it. It will look at how civil society actors like charities deploy online campaigning, how journalists can engage voters more profoundly and what can be done to give them a more informed and meaningful voice in the election process.


  • What are the values that make for ‘good’ online communication at an election and how can they be embedded in electoral institutions?
  • What can be done to improve the transparency of elections and other political campaigns? How useful is transparency as a regulatory device?
  • Information is supposed to improve politics but what can be done to improve both?
  • What do people need to know in order to effectively fulfil their role as citizens? What do they know?
  • How should we address perceived problems such as filter bubbles, lack of deliberation, the challenge of increasing diversity of sources and perspectives, the problems for mainstream political journalism, the rise of polarisation?
  • How should political parties, government, the news media and platforms change to address the perceived decline in trust?
  • What is the information crisis doing to the quality of debate and how can it be improved?
  • Can we shift the focus from what should be restricted to capacity-building? How do we reclaim the original idea that communication technology can be empowering for citizens?
  • To what extent are contemporary challenges to democracy routed in the political system or in the media? How do the two spheres interact with each other?

Have your Say

If you have information for the Commission to consider, submit via our online form