Is It Worth It? An Experimental Examination of the Added Value of Deliberation in a Direct Democratic Process

Abstract It is often argued that deliberation could improve citizens’ acceptance of the outcomes of direct democratic decision-making. However, the available scientific evidence remains limited and it remains unclear to which degree these findings can be generalized and reflect causal effects. We therefore use a randomized survey experiment on a large-scale representative sample of the Dutch population to disentangle the impact of direct democratic processes in the form of referendums in isolation compared to situations where deliberation is added, and wherein the outcome of this deliberation is either (a) congruent or (b) not congruent with the referendum outcome. We find a positive significant effect among our respondents when there is congruence between the deliberative mini-public and the referendum outcome and a negative significant effect when there is incongruence. Both effects seem to cancel each other out: overall we find no clear evidence in favor of an average positive or negative impact of deliberation added to a referendum. In an explorative analysis we find some evidence suggesting that outcome acceptance of incongruent processes is lower the less respondents deem the procedure fair (moderator effect). Lastly, while not designed as direct replication, our study provides some evidence in line with a prior seminal study by Germann et al. (Political Studies 72(2), 677–700, 2024) regarding higher outcome acceptance, specifically among decision losers, when deliberation is added to referenda, although only when the outcomes of deliberation and the referendum are congruent.