Skip to main content
Open Access Publications from the University of California

UC Santa Barbara

UC Santa Barbara Electronic Theses and Dissertations bannerUC Santa Barbara

Presence and Impotence: The perils of guaranteed descriptive representation

  • Author(s): Allen, Geoffrey
  • Advisor(s): Stoll, Heather
  • et al.

Around the world, states are grappling with how to integrate minority communities into social, economic and political life. Increasingly, calls have come from academics and minority community leaders for the creation of guaranteed descriptive representation institutions, designed to secure permanent descriptive representation for marginalized minority communities in the legislative branch. Communally reserved legislative seats are one of the more commonly used institutions to provide such guaranteed representation. Under this system, legislative seats are set aside for the representation of minority interests.

This system is now found in more than 20 states around the world, with many of these states adopting the system after the end of the Cold War. To date, most research on this type of guaranteed descriptive representation institution has utilized single case-studies. One of the most consistent findings in this research has been that communal reservations seem to have little to no impact on the policy-preference attainment of minority communities. This finding contradicts a theoretical literature that argues that providing a political presence for communities should translate into increased policy influence for the targeted communities.

This project posits a theory as to why there is a discrepancy between theory and evidence in this case. I argue that, rather than an anomaly, the lack of policy-influence for communities provided with communal reservations is a feature of the system. The creation of special, minority-targeted electoral districts I believe causes a bifurcation of the political system. Instead of having a national political conversation about ethnicity and difference, the creation of communal reservations allows mainstream political actors to essentially ignore minority concerns, passing them off as the domain of minority representatives alone. The electoral incentives associated with campaigning for minority voters are minimized as a result of the special constituency.

In order to justify the hypotheses I establish about the relationship between communal reservations and policy-influence, I conduct a plausibility probe in the state of Croatia. Using a combination of elite interviews, media analysis and electoral analysis, I show that, at least for within Croatia, communal reservations seem to be limiting the policy influence of minority community leaders, as I predicted. One of the key causes of this, according to my research, is a decision among mainstream political parties and actors to remain effectively silent on minority issues. To establish the generalizability of the theory, I conduct two tests. First, I provide evidence that shows that, systematically, turnout in communally reserved districts is substantially lower than the national average, which I argue indicates low levels of engagement and/or satisfaction among minority community members. Next, I look at coalition participation rates among small parties in Europe, and find that ethnic parties elected through reserved districts are substantially less likely to participate in governing coalitions than other parties, even accounting for party size. I believe this is because the design of communal reservation systems creates disincentives for coalition-building with ethnic minority parties.

This research has two major implications. The first is that, as an tool for providing representation for minority communities, communal reservations are a poor choice. If anything, this institutional design may exacerbate inter-ethnic tensions. More broadly, the findings suggest that, while descriptive representation may provide benefits, guaranteeing descriptive representations may create negative externalities for which at present we have not accounted.

Main Content
Current View