Election Cycles and Ethnic Conflict, page 1
Only recently have scholars started to take seriously the variations that exist among the
different forms of ethnic conflict.
1
Horowitz (2001), for example, recently dedicated a book to
understanding ethnic riots. Fearon and Laitin (2003) focused quite closely on an equally specific
form of conflict—rebellion against the government. And Lustick, Miodownik and Eidelson
(2004) have examined the impact of power-sharing institutions on secessionism.
These and other efforts represent significant progress, suggesting that different forms of
ethnic strife may have distinct causal mechanisms. In this article, we focus on three specific
manifestations of ethnic unrest—peaceful dissent (in the form of ethnically generated protest),
conflict amongst groups, and violence directed against the government. We argue that it is
important to distinguish between these forms of conflict because proposed solutions to any one
type of dissent may actually make another type worse. Thus, we may need to confront
important tradeoffs, such as the possibility that federalism may reduce conflict at the national
level, but increase it at the local level (Saideman et al. 2002). In this paper, we seek to examine
how various democratic structures act to facilitate or inhibit different forms of ethnic conflict.
We are particularly interested in the impact of elections and electoral institutions on
protest and violence. In particular, we are concerned with determining whether electoral
competition exacerbates disputes between groups, and between groups and the government. In
keeping with previous work in this area, we assume that elections, as clearly distributive
institutions, are designed to create winners and losers.
Because the stakes of winning and losing may be very high, it is reasonable to
hypothesize that elections may deepen communal frictions. Tensions may rise, for example, as
balloting approaches and as campaigns mobilize supporters and adversaries. Conversely, friction
may increase as more and more time passes since the last election and grievances accumulate.
By considering the relationships between political institutions, election cycles, and our three
measures of unrest, we can learn not only about the impact of political competition on ethnic
strife, but also about the dynamics distinct to each form of conflict.
This article, therefore, addresses a significant gap in comparative politics literature by
systematically considering the impact of elections and election cycles on ethnic conflict. To this
point, there have been no comprehensive studies of this question, although Wilkenson
(forthcoming) has addressed some of these concerns in the Indian context. Because dissent,
peaceful or violent, is an attempt to get grievances heard, and because elections influence who is
likely to be heard, we ought to expect connections between the two.
In the sections below, we first consider possible dynamics that differentiate the various
forms of conflict, producing predictions for the impact of institutions on ethnic unrest. We use
time series cross sectional data, based in part on information collected by the Minorities at Risk
project, to determine which causal connections seem to matter most around the globe between
the mid 1980s and the late 1990s.
2
After discussing our findings, we deal with the implications
for both policy and future research.
Varying Forms of Ethnic Conflict: Risks, Rewards and Violence
While we can think of many ways in which ethnic conflict varies, we focus here on the
form and target of a group’s activity, using a relatively simple cost and benefit calculus to
determine the strategies groups choose. Form refers to whether the behavior is peaceful or
1
For more discussion of this point, see Brubaker and Laitin (1998).
2
For more on MAR, see Gurr 1993, 2000, and
http://www.cidcm.umd.edu/inscr/mar/home.htm
.
Convention