Rethinking corruption: hocus-pocus, locus and focus

Although there has been a significant increase in research on the phenomenon of corruption over the last quarter-century, there is little evidence that this has resulted in effective policy interventions, nor in any significant reduction in its scope and extent. This article argues that three main reasons account for this failure to develop effective anti-corruption measures. First, the dominance of economistic analyses of the role of incentives in decision-making has given rise to proposed institutional fixes that are too abstracted from reality to gain purchase. That dominance was partly prompted by a misplaced assumption that market-based liberal democracies would become the modal regime type following the collapse of communism. Second, an emphasis on the nation state as the primary unit of analysis has not kept pace with significant changes in how some forms of corruption operate in practice, nor with the changing nature of states themselves. Third, different types of corruption are insufficiently disaggregated according not just to kind and form, but also to the locations in which they occur (sectoral, organisational, geographical), the actors involved, and the dependencies that enable them. This reflects an overuse of the term ‘corruption’ in both academic literature and policy recommendations; insufficient attention is paid to what exactly is being addressed and ultimately, the notion of corruption, without adjectives, is a poor guide both to analysis and to policy prescription