It is often argued that national referendums on EU matters can be a cure for the democratic deficit of the EU and its policies. But what can we learn from a country like Switzerland about how and when direct democracy works? Eva Thomann (University of Exeter), Isabelle Stadelmann-Steffen (Universität Bern), and Eva G. Heidbreder (Universität Magdeburg) conclude that referendums in the EU usually lack the necessary institutional and administrative links between direct and representative decision-making to have legitimacy-enhancing effects.

Today, many perceive the European Union (EU) as lacking in democratic legitimacy. Euroscepticism is on the rise, driven, in part, by the notion that the EU’s electoral and policymaking processes are opaque and distant from national concerns. Some of these perceptions are demonstrably wrong. Still, political leaders in the EU and its member states are aware of this issue and seek ways to reduce their distance to “the people”. One solution – intensely discussed – is that of national referendums on matters of EU-wide relevance, namely national decisions on membership, treaty ratification and specific policy issues. Examples include the round of referendums on the Constitutional Treaty in 2005 in France, the Netherlands, and Luxembourg; the 2016 Hungarian migrant quota referendum; and the infamous Brexit vote in the United Kingdom in 2016.

More legitimate decisions?

The intention behind holding direct democratic votes is often to reach a more legitimate political decision: referendums integrate voters in the decision-making process, and the resulting decisions should enjoy higher levels of acceptance among the voters. People who make this argument often refer to the Swiss political system with its direct democratic elements. In fact, a vast majority of European citizens think that referendums are important for democracy. However, experiences like the Brexit referendum raise significant concerns as to whether national referendums on EU matters lead to more legitimate democratic decisions: the Brexit vote has had a severe and long-lasting effect of politicising and polarising the British voters. It has also raised a host of questions about the rightful place of this vote in the constitutional order of the UK; some argue it has at best an indicative character, whilst others reject any alternative to Brexit as unconstitutional. Having reviewed the extensive research on Swiss direct democracy, we argue that three key lessons can be learned from the Swiss case.

Institutions matter

Everything we know about why direct democracy works in the Swiss political system suggests that it functions well because the direct democratic instruments are embedded in other political institutions. In Switzerland, popular votes are an integral part of the regular policy-making process. Such votes are either constitutionally mandatory or generated by bottom-up forces, that is, by voters collecting a sufficient number of signatures. Moreover, Swiss institutions have a host of safeguards to prevent popular votes from becoming a “tyranny of the majority”. Political decision-making processes are consensual, meaning that different interests, namely the parties, interest groups, the cantons, and the people have their specific role in this process. Switzerland’s federalist structures (the so-called Ständemehr) ensure that the interests of both member states, that is, the cantons, and citizens are equally represented by balancing the representation of the cantons with the direct representation of the people.

Finally, it is important to note that the direct democratic decision is never the final policy-making step. There is actually an – often contested – tendency to adapt unpopular or extreme decisions in the implementation process, that is, to make these decisions more moderate or consensual. All these safeguards serve the same goal: integrating a variety of actors and interests into the decision-making process in order to reach a consensual solution.

In the context of EU member states, these safeguards are by-and-large missing. In EU member states, referendums are typically once-in-a-lifetime votes with an unclear legal status, often leading to fierce contestation. Many national referendums on EU matters are launched by political leaders guided by political, power-related motives. In addition, since referendums on EU matters can only be held on the national level, each EU-related vote is cast by only a fraction of the EU-citizens, despite its results affecting citizens of other EU states who have no voting influence whatsoever. Given that these expressions of direct democracy are currently so dissimilar to Switzerland’s system, can we really expect them to have similar effects in EU member states? We do not think so. When referendums are highly exceptional events and lack a clear institutional embedding that integrates as many affected as possible, they generate uncertainty. Voters are not only uncertain about the outcomes of the vote, they are even unsure about what the government will do with the result. This makes a legitimacy-enhancing effect of referendums not only unlikely; it can invert their intended effect.

Polity must match constituency

If a referendum on an EU policy is held in an EU state, there is an obvious mismatch between the people that vote – that is the people of that country –, and the set of countries, institutions, and people who are affected by the vote. National referendums on EU matters typically have EU-wide implications, even though a large share of those affected are unable to have their say. As the 2016 Hungarian migrant quota referendum has illustrated, even when national referendums appear to concern mainly national issues, they can conflict with the distribution of policy competences between the EU and member states.

Also in Switzerland, the problem of non-matching constituencies exists. In Eurosceptic Switzerland, direct democratic decisions that go against international treaties and obligations are also repeatedly put to vote, sometimes resulting in contradictory democratic decisions with a constitutional or internationally binding character. It is difficult to design stringent rules of the game that deal with such contradictions. In the end, such contradictions challenge the sensitive balance between direct and representative democracy. In Switzerland, such conflicts are sometimes revealed by courts after the vote by declaring a conflict with international law; however, the Swiss constitution does not establish a clear hierarchy between international law and Swiss constitutional law and Switzerland does not have a constitutional court, which makes it impossible to resolve these conflicts through courts. Before the vote, the federal parliament can declare an initiative invalid only if it violates formal rules or the peremptory norms of international law—this happened only 4 times in history. Another, more frequent strategy is, interestingly, to only partially or de facto not implement the popular decision. In other words, while proponents of national referendums on EU matters promote the direct execution of the citizens’ will as the positive example, in Switzerland representative, executive, and judicial institutions often mitigate or even “correct” the citizens’ vote.

The voter side: voting under uncertainty

The outcomes of referendum votes are often difficult to predict and sometimes voting days produce “surprises”. This also pertains to the debate about whether voters are capable of making informed decisions on important issues at all. While some argue that this is actually the case, others emphasise that direct democratic decisions are prone to (emotional) campaign effects and biased information processing. For the case of Switzerland, empirical evidence for both perspectives can be found: voters are often able to justify their decisions, but at the same time, campaign dynamics can influence the vote outcome. Thus, even in Switzerland with direct democracy being deeply embedded in the political process and with voters having extensive experience in making up their mind on policy issues, referendum outcomes are sometimes uncertain or surprising.

We argue that the likelihood for unpredictable and surprising results only increases in national referendums on EU matters. One reason is the above-mentioned uncertainty related to the lack of institutional embedding. Another reason is related to voters’ behaviour, which might be much more volatile if direct democratic decisions are rare events, therefore triggering many different mechanisms beyond the decision as such. Third, the motives for initiating referendums in a national context are often also strongly “internal” national concerns, even though the questions put to vote are EU-wide issues. This mismatch affects both campaigning and the handling of the results. We question whether this inherent uncertainty is conducive to the legitimacy of the political process.

Direct democracy is not a simple fix

Overall, we conclude that national referendums on EU matters are unlikely to have legitimacy-enhancing effects given the current institutional setting. Instead, they are likely to enhance the polarisation and the politicisation of support for and resentment against the EU. The frequently mentioned success of direct democracy in Switzerland is strongly related to the fact that direct democracy is one inherent element of a political system that also includes consensual decision-making, double majority requirements and extensive implementation processes. This does not make Switzerland a better but a different democracy, with its specific strengths and weaknesses.

While national referendums on EU matters ought to improve the democratic quality of EU decision-making, they actually may have the opposite effect. EU referendums, whether at the national level or even if made possible at the EU level, in most EU states lack the necessary institutional and administrative links between direct and representative decision-making. Direct democracy is not just an add-on to parliamentary democracy. To enhance democratic legitimacy, it needs to be carefully crafted as part of representative democracies.

This post represents the views of the author and not those of the Brexit blog, nor LSE. Image C00 Public Domain. It also appeared on EFTA-Studies.org.

Eva Thomann is a Senior Lecturer at the University of Exeter.

Isabelle Stadelmann-Steffen is Chair of Comparative Politics at the Lehrstuhl für Vergleichende Politik, Universität Bern.

Eva G. Heidbreder is a Professorin für Politikwissenschaft at the Universität Magdeburg.