The efficacy of unofficial referendums as regionalisation and independence protest events: issue salience, polarisation, and mobilisation

A significant body of literature suggests that direct democratic processes can have indirect effects on public opinion and, as a result, on policy outcomes in states where such procedures are highly institutionalised. But what about states or regions which have no legal provision for direct democracy? In these polities, direct democracy, in the form of unofficial referendums, is a surprisingly frequent tactic employed by regional groups, social movements, or political parties to influence public opinion and awareness, and thus policy outcomes, towards greater regional autonomy, sovereignty, or independence. Europe alone has seen an explosion in the number of these unofficial referendums over the past decade. There is, however, surprisingly little analysis into the efficacy of these types of unofficial democratic mobilisation, or what effect they might have on public and elite opinion over the medium-long term. I address this gap by offering a comparative longitudinal analysis of the evolution of sovereignty/independence issue salience and polarisation in six referendums across five sub-national regions. These include five unofficial independence referendums in Catalonia, Venice, Lombardy, and South Tyrol, and one official independence referendum in Scotland. Using interrupted time series (ITS) designs in a comparative model, I map the effect of unofficial referendums on issue salience and voter issue positions and compare this to the effect of an official independence referendum and to regional sovereignty movements that have not adopted this tactic, such as in Wales and the Basque country. In doing so, I address the question of to what extent unofficial referendums are an effective means of mobilisation, recruitment, or persuasion for regional parties and interest groups. I find that both unofficial and official independence referendums produce general saliency spikes during and immediately after their campaigns. The extent to which this increased issue salience is sustained post-referendum, however, varies across contexts and may be dependent on the size of democratic mobilisation (referendum turnout), but also on the underlying strength of regional parties and interest groups. I propose a new model for conceptualising the causal mechanisms which might explain the varying efficacy of unofficial independence referendums across different contexts.