Who Should Deliberate? Evidence from a Survey Experiment

Citizens are growing increasingly dissatisfied with democratic decision-making, but there is little agreement on the possible remedies to this problem. Some claim that increasing popular involvement in decision-making is the key (e.g. Barber 1984), while others see a need for greater involvement by policy experts (Hibbing & Theiss-Morse 2002). Another approach focuses on the responsiveness of decision-makers (Esaiasson & Wlezien 2016), which is typically understood in terms of implementing policies that citizens want; i.e. ensuring that citizens get their preferred outcomes. However, others emphasize that decision-makers ought to justify their decisions better to ensure democratic accountability (Warren 1999; Urbinati and Warren 2006). The empirical weight of these factors in shaping citizens’ legitimacy beliefs remains largely unclear. This paper aims to fill this gap. We here examine the impact of these factors on decision legitimacy with a survey experiment conducted on a representative sample of the Finnish population (n=1200). The experiment is related to the on-going discussion around a citizens’ initiative for allowing euthanasia in Finland. The survey experiment includes a vignette where we examine how decision legitimacy is affected by three treatments: 1) What actors are involved: Elected representatives, Ordinary citizens, or Experts 2) The outcome of the decision: Recommendation for or against the approval of Euthanasia (and whether the recommendation corresponds to the preference of the respondent) 3) Justification: Whether or not decision-makers provided arguments for their decision. The experiment thereby has a 3x2x2 factorial design. Our dependent variable is the perceived legitimacy of the decision on euthanasia. We include a number of variables that can affect individuals’ legitimacy beliefs, i.e. those based on actors involved in decision-making (input legitimacy), policy preferences (output legitimacy) and procedures (throughput legitimacy). The experimental design makes it possible to examine various factors affecting legitimacy beliefs pertaining to various decision-making actors and their interactions.