Arranged independence referendums. Explaining the state agreement paradox

This article aims to explain a paradox: why do central governments accept to hold independence referendums if this decision puts at risk their territorial integrity? For this purpose, this investigation tests a series of seven hypotheses on 130 international cases of secession plebiscites from 1944 to 2020 extracted from an updated version of the Contested sovereignty dataset elaborated by Mendez and Germann (2018). The results of the statistical regression identify four relevant variables. In the first place, the competition-proximity model and the quality of democracy explain the support brought by central governments for the organisation of self-determination referendums. Additionally, the study shows that, with time, central governments have become less likely to allow minorities to hold independence consultations. Finally, and with a lower level of significance, it is also argued that periods of previous violence incite host state authorities to agree to hold self-determination referendums.