Economic Growth in Autocracies

Abstract

Previous studies on the effect of political regimes on economic outcomes largely consider autocracies as homogenous regimes. Given the internal heterogeneity in the political institutions and economies of autocracies, using dichotomous classification of political regimes to study their effect on economic outcomes is less informative. This dissertation first decomposes economic growth in autocracies and demonstrates the heterogeneity in the structure and growth of their economies both in distinction with non-autocracies and within the different autocratic regime types. Second, the dissertation addresses a more fundamental question in comparative political economy literature and asks whether autocratic regime types explain economic growth and income level differences across countries and over time. To offer a comprehensive answer, it estimates several static and dynamic panel models for growth rates and income levels. Short and long-run casual relationships are studied using balanced and unbalanced data, across Cheibub et al. (2010), Geddes et al. (2014) and Wahman et al. (2013) autocratic regime classifications over 37 years from 1972 to 2008. Results show that autocratic regime types are not informative in explaining growth differences once time effects are introduced to the model. The study of the effect of autocratic regime types on income levels follows the dynamic panel procedures also used in Acemoglu et al. (2017). Again, the explanatory power of regime types significantly reduces once time effect dummies are included. The results conclusively show that income and growth rate differences are mainly explained by other factors that are common to all regime types. It is also possible that there exists sizable arbitrariness in the way regimes are classified, and political institutionalization within regimes are too diverse that they fail to demonstrate a uniform and consistent effect on growth rates and income levels.