An Exploration of Descriptive and Substantive Representation in South Africa

• In Event: Ethnic Politics

Abstract

Descriptive representation in government has long been assumed to increase the degree to which elected representatives reflect the policy priorities of citizens. However, this literature has rarely looked beyond single-trait studies of race or gender. Citizen and politician identities are complex and intersectional, and we know little about how the interaction of identity traits impact substantive representation. In this paper, we use an original longitudinal study of local councillors in South Africa, combined with citizen level survey data, to explore whether politicians who descriptively mirror their constituents across multiple identity categories provide better substantive representation than politicians who lack shared identity traits. The representative dynamics between local councillors and citizens are particularly salient in the South African case, due to deteriorating levels of citizen trust in local government. We find that greater degrees of trait similarity between citizens and politicians are associated with increased substantive representation on policy issue prioritization, and that the combination of shared race and shared party is particularly powerful in this regard. Additionally, our findings suggest that ascriptive trait similarity (race, gender and language) matter most for increasing substantive representation in the domain of cultural issues, while trait similarity based on life experiences (education and socio-economic status) matters more for increasing representation on material issues. Overall, our findings suggest that increasing descriptive representation is likely to have varied effects on whether citizens have their priorities reflected in the opinion of political elites, suggesting that we need to move beyond single trait studies to better understand this topic.

Authors

• Evan S. Lieberman, Massachusetts Institute of Technology

• Sarah Jane Lockwood, University of Cambridge

• Philip Martin, George Mason University

• Vinothan Naidoo, University of Cape Town