What will drive the outcome of the UK’s EU referendum?

Posted on 29. June 2015

by Christopher Wratil

Christopher WratilDavid Cameron has been returned as prime minister in the UK general elections – not with a landslide victory but with a seat gain sufficient for the Tories to form a single party government. A central electoral pledge of Cameron was his promise to hold an in-or-out referendum on the UK’s EU membership, which he has made a central part of the Queen’s speech outlining the agenda of the new government. It is now crystal-clear that the referendum will be held before the end of 2017 and many people are speculating about the possible outcome of such a referendum. I want to share my perspective here as a political scientist, who studies electoral processes. My aim is not to ‘predict’ the referendum outcome.

I think if referendums, especially on EU integration, have taught us anything, it is that their outcome cannot be predicted years, months, or even weeks before the polling date.Think of the French referendum on the EU Constitutional Treaty that at the point of its announcement in March 2005 was believed to become a ‘cheering for Europe’ exercise. Just within a two-months campaign, a clear majority for the ‘Yes’ side turned into a 55% opposition to the new constitution. Alternatively, think of the Scotland referendum last September, in which polls turned from just about 30-35% support for independence in early August 2014 to 49% a week before the referendum. This does not mean that we do not know anything about the factors that influence people’s vote in such referendums – quite the opposite! The problem is more that the factors influencing the outcome are themselves hard to foresee. What I would like to do here is share with you the factors I think will be important. Hence, I want to outline a checklist of what we have to watch closely when speculating about the referendum’s outcome. The four factors I think might be most important in moving voters on the polling date are the following: people’s preferences on EU membership, the referendum question and offer, the government’s popularity and inter- and intra-party politics. Each one is associated with a high contingency – small things can make a huge difference. However, I start from the least contingent one.

People’s preferences on EU membership

Probably the most stable factor that will determine the referendum outcome are the UK citizens’ preferences on EU integration. With ‘preferences’ I do not mean short-term opinions that may fluctuate starkly but rather a stable long-term baseline of whether or not people like the EU. This baseline is stabilised by social, cultural and political fundamentals. Surveys partly reflect these preferences but not every change in opinion indicates a change in people’s preferences. The graph below shows the lead of supporters of EU membership – those that think EU membership is ‘a good thing’ – over those that think membership is a ‘bad thing’, from 1990 to 2011. The flicking line are the raw responses, the red line is smoothed. This shows that support for EU membership in the UK dropped considerably at the beginning of the 1990s (as in many EU countries). It then remained quite stable between 1998 and 2008, before falling again during the global financial and Eurozone crises. While the question on EU membership is not available for recent years, other questions that are highly correlated show that the preferences of UK citizens on EU integration should be at about the same level as in 2010 and 2011. What is this level? Essentially, it is 0 – there are as many people in the UK who think the EU is a good thing as people who think it is a bad thing. Hence, from the perspective of preferences, this referendum really should be a neck-and-neck race! Can these baseline preferences on integration change before the referendum? The issue is not whether we will see fluctuation in these figures – there always is – but rather whether we are approaching another turning point like the early 1990s (Maastricht Treaty), 1998 (Labour government deciding to not join euro), or 2008 (economic crisis). The most obvious turning points are of economic nature – a sudden end to the economic crisis and strong growth and job creation in the UK or a partial breakdown of the Eurozone. While the first might help the supporters of EU membership, the second would help the opponents.

ChristopherNote: Eurobarometer survey series, author’s own calculations

The referendum question and offer

If people’s preferences determined referendum outcomes alone, it would be much easier to predict them in many cases (certainly not in this case, with almost equally divided preferences over the EU). However, a second factor that will influence the outcome is the substantive issue at stake. Cameron’s portrayal of the referendum as a simple, clear-cut choice of ‘in-or-out’ masks the subtleties of substance that will sway some voters: what will happen in the event of a ‘Brexit’? Will the UK have to leave the single market? Could it stay and, if so, would it still have to contribute to the EU budget (as Norway and Switzerland do…)? Could the fundamentals of the free movement of people be compromised if the UK decided to stay in the EU? We have seen how decisive such questions can get in the Scottish referendum, when in the week leading up to the polls major enterprises announced they would withdraw jobs and investments from Scotland if it broke the union. This certainly impressed many voters. As in the Scotland referendum there will be a public debate, with studies, figures, threats, red lines, and promises. These dynamics are very hard to foresee. However, a key question will be whether the pro-EU coalitioninside the UK will get their act together and whether they are able to coordinate their efforts with the pro-EU coalition in the other member states and EU institutions. My expectation is that threats – especially credible ones – about the consequences of leaving will play a key role in moving undecided voters. Let’s consider the potential potency of promises for a moment and imagine the pro-EU side gets Angela Merkel to support “restrictions on EU migrants’ rights to social benefits”. How credible can such a promise about treaty reform be under the EU’s unanimity and supermajority principles, in which not only a German chancellor but each and every member state parliament and all EU institutions have to agree to bring about change? Paradoxically, a threat by the European Parliament’s President Martin Schulz that the UK will “get no place in the single market” if it leaves, could appear much more credible and help the pro-EU side more. This is simply because if the UK decides to leave unilaterally, a united and opposed European Parliament is all it will take to keep an independent, re-applying UK out of the single market. The EU’s status quo bias will make threats more powerful than promises of change. Hence, people like Martin Schulz who – believe it or not – want the ‘Bremain’ but do not shy away from harsh words, will likely be the government’s key allies abroad.

The government’s popularity

Referendums can often turn into popularity tests for the incumbent government. Like mid-term elections, referendums pose a potential opportunity to citizens to demonstrate that they are dissatisfied with the performance of the government. Jacques Chirac’s defeat in the French referendum on the EU Constitutional Treaty is an obvious example of a referendum in which the government’s (un)popularity played a key role. The impact of popularity is, of course, more likely when the government initiates the referendum and provides a clear cue of what it advocates voters to do. In the context of the UK referendum, Cameron has therefore provided the basic ingredients that could turn the referendum into a popularity test. He will not only generously ‘give’ the referendum to the British people, but he has also made it clear that he wants to ‘win’ the referendum. Two factors that seem important are: first, will Cameron at some point threaten that he might resign as prime minister if he loses the referendum (or at least imply this)? Paradoxically, this could under some circumstances weaken the impact/role of the popularity factor as a defeat of Cameron would then have an immediate impact on national politics – it would not be ‘cheap’ anymore to demonstrate dissatisfaction. Certainly, Cameron will only take this step after the other major parties have explicitly joined him in advocating the ‘Bremain’ Second, will the Tory MPs speak with a single voice or will they be split into a ‘Yes’ and a ‘No’ camp? Cameron has already stated that ministers can either back the government’s line on the EU referendum or resign. Whether he will be successful with this disciplinary measure and how many defectors there will be below ministerial rank has yet to be seen. Again, paradoxically, with a clear visible split within the Tories the referendum could appear less like a test of popularity of the government party and more like a single controversial issue that the government puts to the people precisely because their voters might be as conflicted about the outcome as the party itself. In any case, concerns about his government’s popularity make it likely that Cameron will hold the referendum early on in the legislative term (e.g. in summer or autumn 2016) since we know that governments are typically least popular mid-term (i.e. in 2017).

Inter- and intra-party politics

Finally, the positioning between the major parties and within them will influence voters’ decisions. While we know that politicians send cues and that voters are to some extent receptive to these cues, it is very hard to predict when they are successful. Some research suggests that parties are less successful in persuading their voters of their leadership’s message if they are internally split. On the one hand, as some Tory MPs will clearly cause internal division, this might suggest that Cameron will have a hard time convincing Conservative voters of a ‘Bremain’. However, as we have seen above, internal splits could also make the referendum appear to be more about a single controversial issue than about the government itself. Hence, the net effect the Tories’ internal conflict will have on the outcome is really hard to predict at the moment. It seems more likely that internal divisions within the Labour party will make a ‘Brexit’ more plausible. While Labour has so far strongly advocated to stay in the EU, the new leadership might reconsider this position, weaken it, or simply be unable to discipline diverging groups inside the party. Labour’s ability to persuade its voters is also not guaranteed, since Labour voters are typically more prone to taking elite cues than the Conservatives’ electorate.

In terms of inter-party dynamics one could think that a consensus among the major parties to support a ‘Bremain’ might sway voters to follow this cue. However, referendums like the Danish 1992 failed ratification of the Maastricht Treaty show that even very united political elites (including large parts of civil society in this case) are sometimes not able to convince voters. Indeed, if elites are very united, anti-establishment feelings among the electorate may become more important for the referendum outcome. Considering the previous point, the referendum might not be about the government’s popularity anymore, but about the popularity of the entire political class. Hence, for those who would like the UK to stay, some limited party competition among the ‘establishment parties’ on the issue might be desirable. However, it is not yet clear whether there will be a real competition; Labour will probably convey a clear message for a ‘Bremain’, the Conservatives will be split, and the Liberal Democrats will probably remain pro-European. The more likely scenario is that parties will watch out carefully for anti-establishment sentiment and try to address it by stressing the ‘free choice’ of the voters.

As I stressed from the start, nobody can make any serious predictions about the outcome of the referendum. However, my intention here was to give an overview of what I think could be decisive factors. In summary, this referendum is likely to be a neck-and-neck race – in contrast to all the polls at the moment, which seem to imply a clear majority for staying in. The ‘Bremain’ will be more likely if: the UK and the European economies grow again; the Eurozone problems get solved; if there is uncertainty (with the chance of some painful downsides) about what will follow a ‘Brexit’; the pro-EU camp inside the UK is able to bring in partners from abroad in a credible way; if the Tory government is either popular or (more likely) able to portray the referendum as a substantive question and not a popularity contest; if the Labour party is able to convince its voters, and finally; if all established parties manage to contain any anti-establishment sentiment among the electorate. Certainly, there is a lot to do before the referendum with several predetermined breaking points – but this is what the UK voters opted for. Democracy at its best.