A majority of UK citizens voted to leave the European Union in June 2016. Immediately afterwards, the newly installed Tory government, led by Theresa May, committed to deliver the referendum result and started dealing with Brussels to negotiate a withdrawal agreement.

After almost two years of uneasy negotiations, which resulted in a great deal of criticism and loss of support for Theresa May, on the 13th of November, 2018, she and EU representatives signed a draft deal which has two parts: a withdrawal agreement and a statement on future relations between the UK and the EU.

Almost all of the English opposition parties, together with the Conservative hard Brexiteers, started to attack the draft deal and express their intention of rejecting it in the Parliament vote set for January. Critical positions towards the agreement are many and vary from the economic consequences it would have on trade, employment, and growth, to the tricky issue of the Irish border.

UK parties are proposing some alternative paths. Tory hard brexiteers claim that May’s agreement gives too much control to Brussels over UK affairs and they would rather exit the EU with no deal. On the opposite side, many MPs across different parties support the people’s vote campaign; they intend to vote against May’s proposal in Parliament and call for a second referendum on the final Brexit deal. This idea has also been recently fuelled by the ruling of the European Court of Justice on the possibility that the UK could unilaterally revoke Article 50 and stop Brexit. In between these opposing views, there are many MPs, mainly from the Labour Party, calling for a general election, in the hope of a renegotiation of the deal, even if Brussels has made it clear that there won’t be any. Last but not least, there is increasing support, also among Cabinet members, for the Norway-plus option, which would allow the UK to stay in the European Single Market, while accepting the EU principle of freedom of movement.

The debate around the Brexit deal is complex and heated and, as is often the case with such significant topics, it is just as lively and multifaceted in its online dimension. EuVisions tried to capture this dimension, focusing on the Twitter communication of UK Members of Parliament – the people who will ultimately decide the fate of this deal.

The size of the online debate

We collected all the tweets sent by current members of the UK House of Commons between the 13th of November, when May presented the draft deal, and the 13th of December. We also collected citizens’ reactions to the MPs’ tweets, in the form of replies and retweets. The result was a total of 92,207 MP tweets, and 1,735,406 retweets and replies by the public.

“In the run up to the January vote, the public debate about the Brexit draft deal will probably be reshaped, new elements and issues will emerge, and the voting intentions of parties may change. However, both the offline and the online discussions are by now suggesting a parliamentary rejection of May’s draft deal.”

As a first step, we isolated the conversations about Brexit and the agreement from all other topics addressed by the MPs. We identified 23,599 deal-related tweets, amounting to 25.6% of the overall tweets by MPs. We then looked at the distribution across parties of deal-related messages. Almost half of the tweets were sent by Labour MPs, suggesting a commitment of the first opposition party in Parliament to address online discussions about May’s proposal. Conservatives came second with 32% of the messages.

Parties’ sentiments towards the deal

As a second step, we investigated the content of MPs’ messages about the deal, in order to check whether their online stances towards the agreement reflected the offline intentions to vote declared by the parties.

By means of machine learning algorithms, we separated tweets expressing negative opinions towards the draft deal from tweets either supporting the agreement or simply not expressing an opinion about it (other messages). Out of 23,599 deal-related tweets, 13,516 (57.3%) expressed negative stances towards May’s withdrawal proposal, outnumbering the other messages and representing the MPs’ majority position. We then looked at the negative/other distribution per each party’s conversations about the deal and plotted the results, as shown in Figure 2.

Figure 2: Distribution of negative tweets across parties

Coherent with the offline debate, those parties (five out of eight) which sent a majority of critical tweets about the deal are the same ones that declared their intention to vote against it. Among the parties that sent a majority of positive/neutral tweets, there are some peculiarities worth noting. The Democratic Unionist Party is supposed to vote against the deal, even if it joins the Conservatives in a coalition government; this delicate position could perhaps explain the almost half-and-half division between positive/neutral and negative tweets. From another perspective, the majority of the Conservatives are probably going to back their leader’s proposal, even if the hard brexiteer wing is dividing the Tories in both the offline and online arenas. Finally, Sinn Féin emerged as the party with the highest percentage of non-negative tweets, maybe in relation to its members’ intentions to abstain from voting.

Citizen engagement

After demonstrating the orientation of MPs’ tweets about the deal, we tried to outline their capacity to trigger a reaction from citizens. We built a standardised “index of reaction”, indicating how much more/less a deal-related tweet sent by an MP resonates in other users’ messages when compared with a generic MP tweet. We also compared the resonance of a negative tweet about the deal with the resonance of non-negative tweets. A value of 1 (black-line in the figure 3 below) indicates that there’s no difference in terms of resonance between the compared sets of messages.

Figure 3: Standardised index of reaction around the draft deal issue

As Figure 3 shows, for almost all parties, MPs’ messages about the draft deal have been retweeted more actively than generic ones, suggesting a strong citizen participation in this debate. A deal-related message was retweeted twice as often as a generic one among the Conservatives, and nearly as often for the Liberal Democrats (2.2) and the Scottish National Party (1.8).

Comparing the resonance of the negative set of tweets to the non-negative, we find a more balanced situation. The supporters of the Liberal Democrats and Plaid Cymru retweeted MPs’ deal-negative tweets nearly as often as deal-others tweets; and almost the same is true for the Democratic Unionist Party and the Labour Party.

Interestingly, among followers who most actively retweeted negative messages, we find the Conservatives, suggesting that the supporters of May’s party may be slightly in favour (at least on Twitter) of a no-vote for the deal.

Finally, Sinn Féin followers emerged as the least interested in the draft deal debate; they were the only party that more actively retweeted neutral or positive messages towards the deal than negative ones, perhaps following the neutral line of action declared by their MPs.

Conclusions

In the run up to the January vote, the public debate about the Brexit draft deal will probably be reshaped, new elements and issues will emerge, and the voting intentions of parties may change. However, both the offline and the online discussions are by now suggesting a parliamentary rejection of May’s draft deal. Even if the UK Government is still campaigning in support of the PM’s agreement, it is also preparing the country for an exit from the EU with no deal which, as many commentators have claimed, could be the most dangerous scenario.

Photo credits CC Flickr: UK Parliament feat Alexander Damiano Ricci