Authoritarian order as an equilibrium outcome of distributional battles in politics: The logic of war and political collusion in 19th and 20th century Mexico

What are the foundations of political order? What are the conditions under which democracy and authoritarianism can be sustained? Political science does not have a systematic understanding of order. This dissertation takes two steps towards improving the situation. First, It builds a formal theoretical framework to answer the opening questions. The framework is composed of a family of formal bargaining models that specify the conditions under which war, democracy, and different forms of authoritarian order are self-enforcing. The models are built upon the notion that the existence and type of political order is determined by the way in which powerful actors in a polity solve key distributive conflicts. Authoritarianism is formalized as an agreement among political actors on a distribution of the pie of rule (set of elective public jobs valuable for their policy and patronage attributions); war and elections as outside options. The models highlight how, from the point of view of political actors, both war and elections are costly ways to distribute the pie of rule and therefore collusion is compelling. When costs make collusion unappealing, the outcome is determined by the median faction's favorite outside option and by the feasibility of organizing an electoral challenge. The models also highlight the different logics that sustain authoritarianism.;Second, the theory is combined with evidence to develop systematic explanations for two puzzling cases of construction of authoritarianism starting from anarchy Mexico: the Porfiriato, and the PRI hegemonic-party regime (in 19th and 20th century, respectively). In both cases evidence shows that the expected value of the pie under collusion started being too small to sustain it. Anarchy ensued because most factions in society preferred war to elections. What tipped the equilibrium towards unchallenged collusion was different for each case. In the 19th century an exogenous change---the unprecedented availability of foreign risk capital---generated a paramount increase in the expected value of collusion. Authoritarianism in 20th century Mexico developed through the cumulative effect of endogenous changes that affected the institutions within which successive rounds of bargaining took place. These cumulative changes eventually made violent and electoral challenges more costly than collusion.