Be careful what you wish for: Brexit and the call for another referendum

Whether or not to hold a referendum on Brexit is a clear dividing line between parties in the upcoming UK general election. However, Philipp Harms and Claudia Landwehr argue that support for such a measure is often largely contingent on expected outcomes, and so can entrench political divides. More deliberative democratic innovations might therefore be better suited to resolving the UK’s political conflicts.

People’s Vote march for a Brexit referendum: Picture: Steve Eason / (CC BY-NC-SA 2.0) licence

By any account, British democracy is in deep trouble after the Brexit vote and the failure of Parliament and two Prime Ministers to deal with its result in a way that pacifies the deep conflicts that gave rise to it. But most other seemingly consolidated democracies, too, are confronted with problems of political alienation, growing political inequality and the rise of populist parties and candidates. These apparent ailments of democracy prompt calls for democratic innovations to cure the disease. Democratic innovations, such as deliberative assemblies, participatory budgeting or citizen consultations, can complement or partly even replace representative institutions in some of their functions. Whereas political scientists and theorists of democracy tend to favour deliberative innovations like the latter, the most popular innovations with the public are forms of direct democracy.

In the UK, however, direct democracy – or more precisely, the Brexit referendum – is precisely what caused the present crisis. Scared by the success of Nigel Farage’s UKIP and unable to arbitrate the divisions within the Tory party, David Cameron called the referendum in 2016. Cameron was hoping to put an end to the debate about the UK’s membership in the European Union and tried to use the public as a kind of umpire of last resort, erroneously confident that the umpire would decide his way.

Despite the experiences with the 2016 referendum and its consequences, the People’s Vote campaign as well as the Labour Party and the Liberal Democrats continue to advocate yet another referendum on the UK’s future relationship with the EU. They thus seek to resolve the crisis with the same means that caused it. But will this work? We want to highlight two interrelated and severe problems with the use of referendums to resolve deep political conflicts. Both problems concern referendums in general, but have particular relevance for the British case.

The first problem is that politicians’ motivation in calling a referendum, but also citizens’ preferences for referendums, are insufficiently understood. Referendums are typically justified on the basis of their supposed intrinsic merits for democracy: they are viewed as giving citizens a say on important policy decisions, allowing them to exercise their democratic autonomy more directly, and they promise decisions with a strong public mandate. Asked whether they would support more referendums to be held in their country, citizens will tend to answer on the basis of an assessment of the democratic merits of direct democracy.

As our research indicates, however, the intrinsic motives for supporting referendums are only one part of the picture (Harms and Landwehr 2019a, 2019b). As soon as the decision to hold a referendum is contextualised, a different set of motives comes into play. Where the question is not whether there should be more referendums in general, but whether a referendum should be held on a specific issue, support for it becomes contingent on its expected outcome effects. Put somewhat bluntly, people tend to be more supportive of a referendum if they expect a majority to share their position on an issue. In this case, the referendum procedure may be expected to be instrumental to the achievement of one’s own desired policy preference.

Although a general preference for referendums, which captures intrinsic motives for supporting them, remains an important predictor for choosing a referendum as a decision-making procedure for a specific issue, instrumental considerations thus play an important role. What we can show for ordinary citizens is likely to apply to politicians and office-holders, who are in the position to call a referendum, to an even stronger extent. Placed in a context of strong interdependence, politicians are bound to act and decide strategically and to base their procedural choices on strategic considerations. Referendums are thus likely to be called in the expectation of an outcome that is in keeping with a politician’s own policy preferences and strategic advantages.

At the same time, citizens’ and politicians’ estimation of the majority opinion may well be wrong. In our sample, participants favouring anti-immigration policies (the issue under consideration in our research) tended to assume a majority to share their position and were accordingly supportive of a referendum on the matter. The support for their own position, however, was much weaker than they assumed – psychologists call this phenomenon the ‘false consensus effect’. The significance of instrumental, rather than intrinsic, motives for supporting referendums and the fact that support is contingent on potentially faulty assumptions about congruence between own and majority preferences leads to a further problem.

This second problem consists in the fact that where referendums are called and supported under wrong assumptions about the majority position, their results are unlikely to be accepted as legitimate and decisions based on them may not be expected to resolve political conflicts. A referendum cannot replace participatory, informed and deliberative decision-making processes, but constitutes what Cristina Lafont calls a ‘procedural shortcut’ that will ultimately miss the goal of a democratically legitimate and broadly accepted decision. The Brexit vote is in fact a prime example for a case in which a decision that was based on such a shortcut rather than on inclusive and informed deliberation has failed to settle a political conflict. Instead, the referendum has not only deepened divides, but has also caused severe damage to the representative institutions that are to implement its result.  

Although it remains to be seen whether or not another public vote on EU membership will ever be held, the December election is widely viewed as a quasi-referendum on the terms and conditions of Brexit. Given the first-past-the-post electoral system with its incentives for strategic voting and resulting disproportional representation, however, things are unlikely to improve whatever the result will be. Instead, trust in democratic institutions and procedures is likely to be further diminished. 

Under these conditions, many have called for a deliberative citizens’ assembly to address the issues Parliament has failed to resolve. Given the growing frustration with representative institutions and alienation from politics, however, deliberation should not stop at the Brexit question. Instead, it could take the Irish Constitutional Convention as a model and address democratic decision-making procedures and their innovation as such, aiming to renew the procedural consensus that enables democracies to deal with deep, substantial conflicts peacefully and constructively.

This article gives the views of the authors, and not the position of Democratic Audit.  It draws on the authors’ article, ‘Preferences for Referenda: Intrinsic or Instrumental? Evidence from a Survey Experiment’, published in Political Studies.


About the authors

Philipp Harms is professor of international economics at Johannes Gutenberg University Mainz.

Claudia Landwehr is professor of public policy at Johannes Gutenberg University Mainz.


Similar Posts