By What Authority? Uncovering the Conditional Mandate from Referendums in EU Membership Decisions

A virtue of democratic decision-making procedures is that the foundations in which decisions are made are thought to enhance the legitimacy of the decision. We study the nuances of this assumption and ask under what conditions democratic decisions are seen as legitimate in the eyes of the people. Specifically, we investigate what mandate citizens award an EU membership referendum in three European non-member and three member countries based on the level of turnout, size of majority, and favorability of the outcome. Direct democracy has become increasingly important in the political decision making process on European issues. Indeed, European integration issues of membership, key policies, endorsement of treaties and constitutional documents are the most voted-on issue in the world. As part of the 2017 European Internet Panel Study (EIPS), the survey embedded conjoint experiments are fielded in Norway, The Netherlands, Iceland, Sweden, Germany, and France. EIPS is a collaboration between probability-based, online survey panels. The recruitment strategy of the panels ensures that all citizens have a close to equal, non-zero, likelihood of being invited to participate, ensuring high quality samples representative of the populations of the respective countries. The total number of respondents amount to approximately 17 500 citizens. Utilizing a conjoint experimental design, we propose to the respondents an imagined scenario where European Union membership is put out for a referendum. They are given different scenarios, where turnout level, size of majority, and outcome are varied; three critical dimensions that always will be present in any referendum. Do people think that the government should follow any result once the issue is decided on in a referendum, or do they change opinion when learning about the attributes of the specific referendum? The point of departure is the assumption that people think the results of a referendum on EU membership should be followed, regardless of the turnout level, size of majority, and the outcome of the referendum itself. The alternative hypotheses stipulate that there is a causal relationship between the legitimacy of a referendum and its level of turnout, size of majority, and favorability of the outcome. The experimental design allows for comparisons of ex ante and ex post assessments of the legitimacy of a referendum, as some respondents are not exposed to the treatments before they are asked to assess whether the outcome should be followed by the government.