3 Facets of Political Inequality: Evidence from Citizen Consultations in Kampala (Pre-Recorded)

• In Event: Facets of Inequality

Abstract

A large body of research examines levels and causes of economic inequality but political inequality is much less well understood. This phenomenon can take the form of gaps in citizens' voice (“input” inequality), in the degree of responsiveness of political systems (“throughput” inequality), and in the ways political decisions favor different groups of citizens differently (“output” inequality). Though these dimensions have frequently been assessed in isolation from each other, and sometimes linked with each other, the majority of analyses have relied on observational data. Furthermore, the dimensions have rarely been assessed in the same setting with the goal of disentangling them, and studying their inter-dependence. We seek to disentangle these dimensions of political inequality in the context of a set of citizen consultative meetings organized as part of a process to construct a "Citizens' Charter" in Kampala.

For this, we leverage an experimental design that affords us the possibility to tightly control and measure opinions and behaviors at all stages of this small-scale political process, while at the same time minimizing self-selection issues. Two thirds of 2,312 potential participants in our study were invited to attend 188 small-scale consultative meetings carried out throughout the city of Kampala, where citizens could give their opinion regarding the design of the Charter. Half of the meetings were randomly assigned to be led by local government officers or by nongovernmental third parties. Working in this context allows us to triangulate measurement and to capture the preferences and behaviors of citizens and political elites both before and after the meetings themselves. We collect data on pre-meeting preferences of citizens and meeting leaders, attendance and behavioral data data during the meetings, as well as collective decisions for each discussion topic covered during the meeting.

We find clear evidence of stark, measurable input inequality with, for instance, men and Luganda speakers engaging at significantly higher rates than women, or participants who are non-native speakers of Luganda. There is somewhat weaker evidence of throughput inequality. There is clear evidence that the views of discussion leaders have a strong influence on the outputs of discussions, though the city authority is largely even-handed in whose preferences get heard. Lastly, and most optimistically, we also find there is little evidence of output inequality, at least in terms of the representation of views. The results highlight the independence of three easily conflated dimensions and highlight the scope for ensuring equitable outputs despite inequalities in inputs.

Authors

• Constantin Manuel Bosancianu, WZB Berlin Social Science Center

• Ana Garcia-Hernandez, NOVA SBE & WZB

• Macartan Humphreys, Columbia University