Desktop version

Home arrow Political science arrow Evaluating Democratic Innovations: Curing the Democratic Malaise?

Direct democracy

The widespread diffusion of direct democracy is uncontested. Over twenty improvements for direct democracy took place in the Organisation for Economic Co-operation and Development (OECD) countries within the recent decades (Scarrow 2006). Direct democratic procedures found their way into almost all new constitutions of the post-socialist states, with Lithuania, Slovakia and Hungary granting extensive legislative competences to its citizens, and the Czech Republic, Bulgaria and Rumania the least (Walter-Rogg 2008: 253; Scarrow 2006: 58). Today, sixteen European nation states have institutionalized obligatory referenda in their national or regional constitutions - with some variations. In most states, a referendum is only mandatory if the constitution is to be revised completely or in its main parts, whilst in other states smaller changes require a referendum; and in some states the constitutions demand a referendum in specific cases. For example in Switzerland, Ireland, Liechtenstein, Italy, Denmark and many of the post-communist democracies, the government must hold a referendum on several ‘mandatory topics’. Often, a referendum is demanded before the nation state can join major international or supranational organizations, or ratify a supranational treaty. For example, the Irish referenda on the European Treaties were constitutionally mandated.

In many European states, citizens may file petitions at the national or subnational level.8 Several countries provide initiative rights only at the national level,9 while other countries allow citizens’ initiatives only at the sub-national level.10 In Greece, France and Ireland a popular vote can be launched by citizens neither at national nor sub-national level (Walter-Rogg 2008: 255).

Almost no states have the same set of rules concerning direct democratic procedures. For example, the requirements for a popular vote to be binding differ vastly. In Hungary, for a binding popular decision, at least one quarter of the electorate must take part and at least a simple majority of the votes must be in favor of the petition. In Poland, a referendum is only binding if more than half of the electorate takes part in the ballot. In Slovakia, the hurdle is even higher. Half the electorate must have cast a vote in favor of the petition.11 The role of the government is another example of the differences in the rules. Whereas in Ireland the government must keep a neutral stance and is forbidden to spend money to support one side, many countries do not have such strict rules. Also, the amount of elite control concerning citizens’ petitions varies largely. Only in Italy, Slovenia and Lithuania can citizens initiate a popular vote without the approval of the government or the parliament (see also Kriesi in this volume). So, the sheer existence of direct democratic options does not tell much about the real options.

Looking at the usage of direct democratic options at the national level, Switzerland and Liechtenstein are leaders. They are followed by Italy, which has had 87 referenda since the introduction of this option. France (38), Ireland (29), Denmark (19), Lithuania (18) and Spain (16) are next on the ranking list (Walter- Rogg 2008: 246).12 Most of these direct democratic procedures have taken place in recent years. This rise was sparked partly by the use of citizen-initiated direct democratic procedures: more than 500 petitions were submitted between 1991 and 2000, which is more than twice the number of any previous decade (see for details Scarrow 2006: 50 ff., 58). Another factor adding to the growth is the European Union. No other issue has activated as many direct democratic procedures as the European Union (EU). Between 1970, when the French voted on the enlargement, and 2005, almost fifty popular votes had been conducted on EU memberships, EU treaties or the euro.

Direct democratic options and usage have not only increased at the national level, but also at the sub-national level (local, regional) (Scarrow 2006: 50): many European states allow local popular votes. They are binding in many post-socialist states (such as Slovakia, Poland, Hungary) and in Germany, but in the majority of states they are only advisory.

Research about the usage of direct democratic procedures at the local level is a difficult task, because data are not easily available. What can be said is that - unsurprisingly - the usage depends mainly on the hurdles. In Germany, for example, the number of popular votes varies significantly between the German member states (Mittendorf2008: 81). Bavaria, which has the lowest hurdles, is the most active state, and in states with higher hurdles, fewer referenda are initiated. However, the numbers do not tell the whole story. Just the threat of a petition often forces local representatives to take citizens’ preferences into account and to act in a more responsive way. Direct democratic options often became a striking ‘sword of Damocles’ (see also Kriesi in this volume).

What is known about the impacts of the direct democratic procedures concerning the criteria for evaluation described above? Considering the criteria of inclusive participation, which strata of society cast a ballot? Not many studies are available, but an analysis of German local popular votes reveals interesting patterns (Gabriel and Walter-Rogg 2006: 48). In contrast with most forms of participation, men and women engage to the same degree and there are no significant differences between income groups, though the level of education shows the ‘normal’ participation bias: people with higher education are more strongly involved. However, people with lower education are more often involved in direct democratic activities than in political party activities (see also Kriesi in this volume).

Perceived legitimacy seems to be higher in states with comprehensive direct democratic options. Citizens of these states are more likely to regard their political system as legitimate (see Kriesi in this volume). Whereas, for example, 88 per cent of the Swiss are politically contented, the European average lies between 50 and 60 per cent (European Social Survey 2004; Bowler and Donovan 2002; Lindner 1999).13 But trust in politicians and the government is not necessarily improved by the use of direct democratic procedures (see, for example, Gilens et al. 2001). Again, Germany can serve as an example: German citizens, who participate in popular voting, do not necessarily trust politicians more than non-active citizens (Gabriel and Walter-Rogg 2006: 49).

The deliberative quality in the context of direct democratic procedures is, on the whole, rather low (see LeDuc 2006). Direct democratic options do not necessarily lead to well-thought deliberations about the different options. Most citizens stick to ‘heuristic cues’, as Kriesi has already pointed out in this volume.

Without doubt, direct democratic procedures help to identify collective goals: they inform about the preferences of the majority of the voters. Direct democracy also seems to support the achievement of collective goals more often than not. States providing more direct democratic options seem to achieve a better performance than those with fewer options - at least in the Swiss case (Moeckli 2007: 122). Swiss cantons with more direct democratic options have lower debt rates (per capita), they provide services more efficiently, and their decisions are more often in compliance with the preferences of the population. They also score better when it comes to the fulfillment of welfare-state policies (Vatter 2006; Kriesi in this volume).

In relation to the improvement of political knowledge, most studies show that citizens are better informed about politics when they have direct democratic participation options. For example, citizens in states with EU-related referenda know more about the EU than citizens in states without EU related popular vote (Benz and Stutzer 2007; Kriesi in this volume). However, direct democratic options do not necessarily improve participation in elections: they neither lead automatically to higher nor to lower voter turnout.14 Considering the improvement of civic skills, comprehensive direct democratic options seem to improve internal efficacy (for example, see Tolbert et al. 2003). But until now, not much is known about the impacts on qualities such as tolerance or the ability to compromise.

Table 8.2 summarizes the impacts of direct democratic procedures.

Table 8.2 Evaluation of direct democratic procedures

Input-legitimacy Inclusive equal participation Perceived legitimacy


+ (political system)/

0 (political representatives)

Democratic process

Deliberative quality



Identification of collective goals


Achievement of collective goals


Civic education

Improvement of knowledge


Improvement of civic skills


174 Brigitte Geissel

Number of participatory budgets in Europe by country (1993-2005)

Figure 8.1 Number of participatory budgets in Europe by country (1993-2005)

Source:, May 2008.

< Prev   CONTENTS   Source   Next >

Related topics