Populating the Democracy Cube: Assessing Power Delegation, Participant Interaction and Representation in Participatory Governance

Since the publication of Archon Fung’s 2005 paper on “Varieties of Participation in Complex Governance”, the model of the “democracy cube” has led researchers internationally to assess the intensity of participatory governance in the dimensions of (1) power delegation, (2) participant interaction and (3) representation. This research contributes to the debate on assessing participatory governance in two ways: First, we discuss the dimensions of the democracy cube as to its empirical applicability, with a special view to whether and how each dimension constitutes a quantitative scale in the sense of “more” or “less” intensive participation, also drawing on recent literature that explicitly tests the model. Second, we perform a quantitative “test” of the model by drawing on a meta-analysis of 305 cases of (more or less) participatory governance from 22 countries in three continents To this end, each dimension was operationalised into one or more different items that were coded by three independent raters for each case. Also, more specific data on the participatory processes at hand, their contexts and outcomes were coded. Perhaps most strikingly, we find the three-dimensional space as defined by the cube to be hugely unevenly populated. In particular, the dimensions of power delegation and participant interaction are highly correlated such that few cases exist which are high in power delegation but low in participant interaction and vice versa. When comparing average values of power delegation and participant interaction for different participatory formats (such as hearings, public meetings, roundtables or public referenda), those two corners of the cube remain virtually unpoluated. This raises questions about the independence of the two dimensions. The dimension of representation – who sits at the table – is most difficult to operationalize in a quantitative “less and more” fashion. Therefore, we defined a number of different items such as the number of participants, the share of participating professionals, and the mode of participant selection (targeted versus open). While some of these also shows a certain correlation with the dimension power delgation, the space is more evenly populated here. We conclude by discussing the usefulness of the democracy cube model for assessing participatory governance and propose suggestions for improvement.