Varieties of Corruption : The Organization of Rent - Seeking in India

How is corruption organized? Studies of corrupt behavior to date shed light on both the causes and consequences of corruption. Yet we have little understanding of how corrupt activities are structured and the ways in which rents are, or are not, distributed across various actors—insights that would, in theory, prove enlightening for efforts to reduce corruption. In this paper, I analyze the organization of corruption through a set of related questions: Are rents from a single bribe distributed across multiple actors? If so, do different types of actors benefit differentially from different types of corruption? What factors, such as the type of corruption or the degree of government centralization, are associated with variation in the distribution of rents? To explore these questions, I first present a new, three-level typology of corruption emphasizing the type of actor paying a bribe and roughly reflecting the character of illicit acts across three realms: high-level policy-making, e.g. bribes for favorable legislation; mid-level policy implementation, such as kickbacks for government contracts; and low-level delivery of public services, for example the payment of “speed money” by citizens. I then draw on new and original data from surveys of Indian politicians to assess how the distribution of rents across actors varies as a function of the type of corruption and the degree of government centralization. I show that there is considerable division of rents across government and non-government actors and the perceived distribution of rents is strongly associated with the type of corruption, though not necessarily in the ways predicted by existing theory. In addition, I find a mixed relationship between government centralization and the distribution of rents. These results validate the utility of a more disaggregated typology of corruption and provide the first clear evidence of the extent to which different political actors benefit from diverse corrupt acts. This draft: April 30, 2013 Prepared for presentation at the Westminster Model of Democracy in Crisis? conference at Harvard University, May 13-14, 2013 Acknowledgments: This research is funded in part by a grant from the Policy Research Institute at the University of Texas at Austin. I would like to thank Bhartendu Trivedi and the MORSEL team for their assistance in data collection. Thanks also to participants in the Yale Program on Democracy workshop and the McGill University Development seminar for comments on earlier drafts.