It is almost too obvious to state, but access to public services and the nature of governance varies hugely within countries, regions and cities. Nevertheless, most work on the “quality of government”, rule of law, corruption, etc. focuses on between-country comparisons. After providing some evidence that within-country variation belies any notion of a national “quality of government”, I lay out a framework for explaining why outcomes vary so much across localities within countries. I explore the usefulness of the framework by providing evidence from three ongoing projects. The first relies on surveys designed to examine the role of slum-level social and political networks in conditioning access to basic public services in Udaipur, India. The second project relies on four post-civil war settings to understand why authorities target some localities with electrification projects but not others. The third project involves a field experiment embedded in an aid program that compares alternative means of improving accountability in Ghana’s district governments. I will conclude with some reflections on the costs and benefits of working with donors on governance programming.
Erik Wibbels is an Associate Professor of Political Science at Duke University. His research focuses on development, decentralized governance and other areas of political economy. He has also spent considerable time working with USAID's Center of Excellence on Democracy, Human Rights and Governance in an effort to improve the quality of aid programs aimed at decentralized governance and service provision.