The Dynamics of Weak Institutions: Rethinking Information and Elite Mobilization in Nondemocratic Politics

Abstract: The paper studies stability and change of authoritarian/weak institutions. In contract to strong institutions, weak ones do not allow binding agreements to be made among players---especially, between the ruler and the ruled---under them. For over a decade, there has been such a vast literature telling us their profound implications for the survival of dictatorships and the making of inefficient economic policies. For example, Acemoglu et al. (2004) explains how a weakly institutionalized environment allows kleptocrats to use a divide-and-rule strategy to exacerbate the collective action problem among potential opposition groups and stay in power. This literature, however, hasn't told us much so far about how and why weak institutions can remain stable or are subject to change. The chief theoretical advance this paper makes to the formal literature on nondemocratic politics is that I characterize the authoritarian/weak institutions as noisy signaling of the information about regime strength. This then allows us to base our new modeling approach on a theoretical finding that an actor's higher-order belief in a global game about others' participation decisions could become non-Laplacian. Owing to the self-enforcing nature of weak institutions, their change and stability clearly depend on whether, collectively, the elite are able to pose credible rebellion threats to the dictator. An important implication following this result is that the variance of one's belief will be increased and, instead of giving all aggregate actions an equal probability, individuals give weights only to the extreme outcomes and therefore are less likely to reach coordination with others to push back the dictator's attempts to change authoritarian institutions.