Ethnic divisions and economic performance

This dissertation studies the consequences in terms of economic performance of politico-ethnic divisions in weak states. In the context of the African continent, where national institutions are often non-democratic, power-sharing arrangements between the elite of the various ethnicities composing society is a frequent mean to achieve social peace, but it has repercussions on efficiency and redistribution. Chapter 1 presents a model of appropriative competition and strategic consumption from a common good, between politically organized competing ethnicities. It sheds light on a mechanism that can sometimes counterbalance the resource curse, depending on the ethnic configuration. An increase in the ease to appropriate common resources can in some cases be favourable for economic growth, when the dominant group raises its influence and investment in response. Chapter 2 presents a theoretical model of coalition formation between ethnicities and studies the effect of the interaction between the quality of institutions and the inclusion of ethnic groups in the central state power on economic growth. It predicts that inclusion is beneficial as long as institutions are good enough. We support this prediction with empirical evidence from panel data estimations for sub-Saharan countries since their independence. In chapter 3, we look at the effect of the political activity of ethnic groups on the economic performance of their related neighbouring kin in Africa, using light intensity measured by satellites as a proxy of local development. Based on a panel dataset constructed from geo-localized information, we find a positive effect, even controlling for spatial interactions. Alongside, we find that ethnic groups represented in the central government are affected positively.