This dissertation studies local governments and the effects of their vertical and horizontal structures on public goods provision in India. The first chapter focuses on political representation and the vertical structure of decentralized governments. Political decentralization combined with minority representation has been purported to give power to the poor. Yet, it is unclear what form of minority representation can best achieve this. In this paper, I ask whether group (mis)alignment across local and intermediate level representations affect public goods distribution to the poor in the context of the Indian National Rural Employment Generation Scheme (NREGS), one of the world's largest social welfare program. Exploiting changes in caste representation driven by India's reservation system intended to increase minority caste representation, I show that minority representation at the local level alone does not increase the transfer of public goods to minority castes. Instead, I find more transfers when there is minority representation at both local and intermediate levels of government. Finally, I show policy-relevant heterogeneity effects coming from electoral motivations of intermediate level representatives and tastes for own caste under a decentralized government. The second chapter examines the horizontal aspect of local governance using India's vastly different rural and urban local government structures. There have been increasing voices that rural local governments lack capacity to govern areas with burgeoning population. I test if this is true and whether local governance affects access to public services, such as treated tap water and closed drainage, in general. To do this, I compare public goods provision between rural and urban local governments after controlling for observables, level of urbanization, and fixed effects. Importantly, I create an objective measure of the extent of urbanization with daylight satellite data and population data. I find that despite the inclusion of these controls and fixed effects, there are positive and statistically significant effects of having an urban local government. I also provide results for placebo tests that show government structure does not directly impact access to private services. Finally, I explore financial decentralization of local governments as a channel.