Compulsory voting, economic conditions and turnout: explaining the outcome of constitutional referendums

Abstract

Referendums were historically and theoretically justified as a people’s veto. Do voters use them as such or are referendums merely second-order votes? This article aims to answer this question through a comparative study of all constitutional referendums around the world between 1980 and 2022 using a VP-Function model. The results indicate that the support for constitutional referendums follows a pattern in which compulsory voting, economic conditions and voter mobilisation are important. Contrary to findings in more generic studies of referendums, there is no indication of a ‘honeymoon’ period for constitutional referendums. Also, in contrast to other studies, the presence of emotive words and what may appear as ‘leading questions’ favour higher support, though this is not present in countries with compulsory voting. The results contribute to the study of referendums and to the wider debate about voters’ preferences by showing that political factors are more important than structural factors.