What will Twitter’s ban on political advertising mean for the coming election?
The December general election will be the UK’s fourth of the social media age, its fifth national poll if we count the 2016 EU referendum – and each election has brought increasing concerns about the power of social media to unduly influence the outcome.
As parties and campaigners have built up a wealth of knowledge about how best to use social media to amplify messages and target certain voters, fears over the impact of micro-targeting, dark-ads and filter bubbles have only mounted. This time around, we can add worries that deep-fakes could be used to deliberately mislead the electorate.
One key concern is that social media targeting allows for messaging to fracture public discourse and debate. Against this backdrop, Twitter has made a welcome announcement: that it will ban paid-for political advertising altogether.
The announcement itself was quite casual, arriving in a Twitter thread from CEO Jack Dorsey. The full policy will be announced on 15 November, with the new restrictions then put in place a week later. And so the implementation of the ban will arrive in the middle of the election campaign.
Still, questions remain. What will this ban look like – and how effective will it be?
The devil will be in the detail, but there are some hints in the initial announcement. Twitter has made clear that this is about issue-based advertising, and not about restricting candidates’ speech. Not only will candidates be prevented from buying adverts; it seems that adverts will be banned if they’re deemed to contain political content.
This is not an easy line to draw. The category of “political” is notoriously difficult to define; everything hangs on how Twitter draws that line and what content it therefore prohibits.
If defined too broadly, then it could stop all sorts of social campaigns from achieving visibility through advertising; if too narrowly, it will easily be circumvented. There are also likely to be contested decisions about what’s banned and what isn’t, with accusations of bias likely to follow.
We might also wonder how the vast knowledge of social media campaigning that’s been built up across three elections and a referendum will inform the use of social medial beyond paid advertising – which, after all, is only part of the picture.
Jack Dorsey’s announcement indicates that the “reach” of political messages “should be earned, not bought”, presumably meaning that it’ll still be possible for parties to circulate their political messages on Twitter provided they find their own ways to spread them.
At this they are increasingly skilled. The party machines understand the effectiveness of political campaigning online, built up with the assistance of data analytics software; they now have a wealth of insights into how to reach an audience and what that audience will like, respond to and share.
In other words, political messages can still reach a wide or targeted audience if you know how to do it. This “organic” reach is powerful and persuasive, and provided they have the right type of insights, any group or individual can find ways of “earning” it.
Still, the step Twitter is taking indicates what might be possible, and how the public sphere might yet be saved from a collapse into opaque targeting. It also highlights that no other major social media platform has made a comparable commitment.
Without Facebook and others joining in, we are still likely to see another election in which paid social content has an unclear yet powerful impact on the outcome.
Nonetheless, Twitter’s move is to be cheered. It will hopefully push campaigning back into the light of day and enable proper scrutiny of the messages. We’ll just have to wait and see how the company enforces the category of “the political” in the first instance – and whether political actors are just too social media-savvy now for the ban to really matter.
David Beer is Professor of Sociology at the University of York. His new book is The Quirks of Digital Culture