The political sphere has been abuzz last week with the announcement by Twitter founder and CEO Jack Dorsey that his platform would no longer accept political advertisements. The policy change will take full effect on 15 November, just in time for the UK General Election in December. It is not wonder that the decision was widely greeted with enthusiasm, both in the UK and internationally. As we have explained here, micro-targeting may have a significant potential to disrupt fair elections - particularly in winner-takes-it-all systems.
Twitter's stance on political advertisement is in stark contrast to Facebook's recent decision to exempt political campaigns from fact-checking, citing freedom of speech concerns. It is easy to see how this is a PR disaster for Zuckerberg and Co, as many users are left wondering why Facebook is still happy to cash in on the growing problem of political disinformation. It seems disingenuous to have entire teams dedicated to stopping bad actors from exploiting their platforms - while at the same time the ad department is happy to help the same actors manipulate users through micro-targeted and often misleading messages. Unfortunately, amidst all the praise (for Twitter) and finger-pointing (at Facebook), some nuance has been lost.
A drop in the ocean
Facebook and Twitter are usually mentioned in tandem when it comes paid-for digital campaigning and by association online disinformation. But taking a closer look at the numbers of the last UK general election shows that, as ever so often in the tech/politics/journalism bubble, Twitter's reach is vastly overestimated.
According to the Electoral Commission, political campaigns for the 2017 UK General Election spend almost £3.2 million on Facebook ads, £1 million on Google ads and only £55K on Twitter ads. Since October 2018, £9.3 million were spent in the UK for a total of over 115,000 political ads on Facebook.
Dorsey acknowledges that Twitter only makes up a tiny share of the entire ad market. The announcement should rather be seen an attempt to publicly distance the micro-blogging platform from the industry behemoth that is Facebook, which has taken the brunt of public criticism over electoral interference over the past months. Banning political advertisement is a principled move to protect 'civic discourse' that, many hope, will force other platforms to fall in line. Twitter leading by example - or so it seems.
Leading with principle or following suit?
The 'principled' stance by Twitter should certainly be taken with a grain of salt. Let's remember that Twitter is not the first major platform to ban political advertisement - in fact, it's one of the few major players that still allow paid-for political campaigns to use their ad services. Microsoft-owned LinkedIn and Bing have both banned political ads over the past two years. TikTok and Pinterest similarly do not allow ads pertaining to political issues on their platform. Reddit, one of the most visited websites worldwide, has banned political ads related to issues outside of the US.
This only leaves Facebook, Google and, until mid-November, Twitter as the major platforms to directly cash in on political advertisement and, by extension, coordinated political disinformation. But even Google has recognised that political ad revenue cannot offset the regulatory and reputational costs in some regions. For example, Google does not allow political ads in Washington DC. The company will also ban political campaigns from their ad platform for the duration of the national election period in Taiwan, following similar approaches to past elections or referenda in Canada, France and Ireland. In India and the EU, political advertisers must now be verified by Google or risk being reported to the authorities, i.e. the Electoral Commission, for non-compliance.
'Organic' traffic, not ads, is the name of the Twitter disinformation game
It all boils down to one simple fact. Ad space may be abused by bad-faith actors, but political ads are not the main medium through which disinformation is spread on Twitter. Instead, it is coordinated inauthentic activity in the form of bot networks, astroturfing campaigns and hashtag-hacking that makes Twitter a cesspool of disinformation. As a myriad of researchers have uncovered during the recent European Parliament elections, Twitter disinformation campaigns relied heavily on fake accounts amplifying divisive and misleading narratives. According to ISD, almost half of the most active Twitter accounts promoting official UK party accounts "show[ed] signs of bot-like hyperactive posting rates". Similarly, we recently discovered a coordinated, inauthentic pro-vaping campaign on Twitter targeting US president Trump. There is little evidence that any of these networks made use of Twitter's ad services. Why pay for expensive advertising space when its far cheaper to purchase a couple of thousand sockpuppet accounts to get your message out?
Twitter's decision is indeed a bold move - it puts significant pressure on the entire ad-tech industry. But how will Facebook and Google react? Yes, public pressure to reform their ad businesses is arguably higher than ever - but Dorsey's announcement also leaves an (admittedly small) gap in an already duopolistic market that Facebook and Google will be happy to fill. Facebook has already admitted it won't fact-check ads by election candidates for the UK General Election in December.
It will be interesting to see the details of the policy change once in comes into effect on 15 November - particularly because important questions are raised about how Twitter will seek to enforce these new rules. Others were quick to point out that political ads are an important messaging tool for insurgent political actors and civil society organisations. Twitter must find a way to ensure their voices can still be heard.
Let's hope that, in light of Twitter's decision, the ad-tech industry will finally be subject to greater public scrutiny. But let's also not forget that ad-targeting is only one instrument in the large disinformation toolbox of bad-faith actors - particularly on Twitter.