Measuring the Quality of Referendums – A Proposal for A Composite Index

Citizens' trust in politics and representative electoral democratic institutions is waning. This does not mean however, that citizens are calling for the abolishment of democracy. To the contrary, citizens are calling for democratic innovations and for democratic institutions that allow for more and more meaningful participation in addition to the current representative electoral democratic institutions. One form of such democratic innovations are popular votes on issues of policy, i.e. referendums. Although referendums are not new, their use has increased in recent years. Today, more countries hold more referendums than in the past, suggesting that they may be one solution to today’s crisis of representative electoral democracy. As with elections however, not all referendums are made equal. Referendums come in many varieties and take place in many different contexts, across democracies and autocracies, in rich and poor countries, large and small. In some contexts, referendums increase the legitimacy of political decisions and empower citizens. In other contexts, they are instrumentalized as democratic window dressing by those in power. We contend that referendums can play an important role in fostering citizen participation and increasing citizen trust, but that not all referendums can fulfil this role. Just as easily, referendums can serve as tools of autocratic rule. In the last decades, the conceptualization and measurement of the quality of democracy has made remarkable progress. Today, we have at our disposition a wide variety of indices and indicators with which to measure and talk about democracy. At the same time, we currently lack a conceptual framework with which to assess referendums – the quality of referendums is undertheorized and underexplored. The goal of our paper is thus to conceptualize the quality of referendums and to develop an analytical framework with which to assess them. To that end, we distinguish between two dimensions of a referendum – context and popular vote. The context dimension captures the institutional setting and environment in which a referendum takes place, whereas the popular vote dimension captures the quality of the actual referendum. On each dimension, we develop a variety of indicators covering the entire referendum cycle. These indicators are then aggregated into the Referendum Quality Score. To conceptualize referendum quality, we build on the normative foundation of the Revised Code of Good Practice on Referendums by the Venice Commission of the Council of Europe and prior research on electoral integrity and referendum integrity. Methodologically, our work is informed by the democracy measurement methods developed by the Varieties of Democracy Project. After conceptualizing and operationalizing referendum quality, we assess the Referendum Quality Score for a number of distinct referendums across time and space to showcase its use. Our aim is to provide a tool with which to distinguish meaningful and democratically empowering referendums from illiberal and oppressive referendums and to thus contribute to the literature on the merits and pitfalls of referendums as democratic innovations.