Essays in the Political Economy of Redistribution and Nation Formation

Abstract

This dissertation mainly explores the effects of income inequality and other forms of heterogeneity —and in particular at the regional level— on the individual willingness to implement general redistribution schemes, and to form part of a political union. In the first part, we introduce heterogeneous social preferences in a standard model of voting on a redistributive parameter in a direct democracy. In particular, and in accordance with experimental evidence, we assume that selfish, rawlsian and utilitarian voters coexist with given proportions. We characterize implicitly the unique political equilibrium of this economy, and prove its existence. It turns out that the level of redistribution in the heterogeneous economy may be either lower or higher than in the selfish one. Furthermore, we show that slight variations in the relative proportion of a given type may lead to very important changes in the extent of redistribution, and we illustrate the implications this may have in the context of the political economy of border formation. Finally, we investigate the theoretical implications of the model regarding the link between inequality and redistribution, and show that it yields different predictions than the standard model with self-interested voters. In the second part, we study the choice between centralized and decentralized redistribution in a political economy model assuming regional heterogeneity regarding both average income and group identity. While centralizing redistribution allows for a potentially beneficial pooling of national resources, it may also decrease the degree of solidarity in the society as a result of group loyalty. In this context, we show that total welfare maximization is closely linked to the minimization of inequality both within and between regions. Analyzing separately two particular cases under direct democracy —no interregional inequality and no group loyalty— we highlight the existence of a scale effect and a pooling effect of centralized redistribution, respectively. In both cases, centralization Pareto-dominates decentralization, from which it follows that the rationale for decentralization only arises when both sources of regional heterogeneity are present. In turn, this means that the Decentralization Theorem does not hold in our political economy approach of redistribution with group loyalty. Finally, allowing for voluntary interregional transfers under decentralization, we show that, due to free-riding, centralization always welfare-dominates decentralization with transfers. Furthermore, it is not even generally true that allowing for such transfers is welfare-increasing under decentralization. In the third and last part, we study the use of partial decentralization as a way to avoid wasteful secessionist conflict in the presence of income disparities between regions. Although decentralization allows regional governments to better target local preferences, which in principle mitigates conflict, it also exacerbates interregional inequality, which in turn may contribute to conflict. We show that, even though an increase in inequality fuels conflict in both regions, the probability of a secession occurring through conflict may be either increasing or decreasing in inequality, depending on whether unifying is socially efficient or not. It follows that, on the one hand, the range of decentralization levels such that the peaceful (decentralized) outcome is politically sustainable always increases with inequality, while on the other hand, the particular level of decentralization that is implemented under the threat of conflict may be either lower or higher. Finally, we show that when decentralization is an irreversible process, it cannot prevent secessionist conflict for any level of inequality.

Date of Award	26 Jun 2012
Original language	English
Supervisor	Caterina Calsamiglia Costa (Director)