Date of Award

May 2017

Degree Type


Degree Name

Doctor of Philosophy


Political Science

First Advisor

Dave Armstrong

Committee Members

Antu Murshid, Shale Horowitz, John Reuter


Conflict, Constitution, Ethnic Outbidding, Ethnic Parties, Ethnic Party Bans, Violence


Political representation through exclusively ethnic parties has long been thought to create, or enforce, social cleavages leading to conflict. To gain support and mobilize ethnic constituents, ethnic party leadership has incentive to exaggerate differences between, or even antagonize, members of other ethnic groups through the process of ethnic outbidding. Classic political theory cautions that the exclusive nature of ethnic parties can also produce a dangerous zero sum game between ethnic groups that cannot be solved by compromise via democratic institutions. Several institutional solutions have been proposed to counter the problem of instability ethnic divisions create for new democracies, encountering varying levels of success. Constitutional ethnic party bans are designed with the intention of coercing ethnic groups to form inclusive, multiethnic political parties. Party membership, then, takes on a national, rather than communal, character, which is thought to prevent extremism and encourage moderate political parties. Opponents of this method of party regulation argue that prohibiting ethnic groups from forming parties obstructs advancement of interests exclusive to the ethnic group and that this practice is especially repressive for ethnic minorities. The theory I present in this dissertation makes the case that ethnic party bans, in the form of constitutional party regulations with spatial distribution requirements, instead prevent majority ethnic groups from resorting to extremism by restricting parties to compete in elections only when the party has diverse membership. To garner support across ethnic cleavages, through a kind of party-level federalism, party leaders are compelled to moderate political platforms in order to form winning coalitions. I argue that this institutional arrangement is not only effective at reducing conflict, but that the design is more stabilizing than power-sharing or proportional representation. To date, there is no consensus on whether the risk of unrest is heightened by ethnic party competition or subsequently dampened by the banning of ethnic party participation in politics. Recent empirical inquiries into the relationship between ethnic parties, or banned ethnic party activity, and conflict have returned contradictory results. Some of the discrepancy in findings can be attributed to differing approaches to research design. In particular, in Ishiyama's 2011 study his mixed findings indicate that in certain statistical models ethnic parties are more prone to conflict whereas utilizing alternative modeling strategies, there is no apparent relationship. I argue that the problem requires a deeper inquiry into measurement attributes of the data which can offer insight into the selection of appropriate modeling techniques. My research deals with this measurement problem by utilizing alternating least squares optimal scaling (ALSOS) regression to effectively transform the ordinal dependent variable to its least biased linear form. As an added attempt to reduce endogeneity, I employ causal inference techniques to stratify a sample of most similar cases between ethnic groups with parties, without parties, and those who have been banned from party operation. This scaling solution will both improve this particular model of ethnic conflict and demonstrate the value of ALSOS regression in myriad social science applications. The analysis follows with discussion of several cases of constitutional ethnic party bans, examining the specific features of the institution that prove most useful.

I find that constitutional ethnic party bans are an effective tool in the prevention of ethnic outbidding. Ethnic groups banned from forming exclusively ethnic parties engage in lower levels of unrest as compared to groups with ethnic parties. I also show that bans with spatial distribution requirements work well to facilitate minority ethnic group representation within multiethnic parties and that bans are less effective in countries with already well-established ethnic parties.