Corruption is an ancient phenomenon, which has attracted increasing attention in recent years. This does not least reflect the growing awareness of the magnitude of the phenomenon. The primary concern of this paper is to discuss the causes and consequences of corruption in developing countries, and possible interventions towards reducing corruption in the public sector. Principal-agent theory (incentive theory) is used as a point of departure. The paper examines the key assumptions of this theoretical approach. Based on empirical observations, it also examines limitations of principal-agent theory for analysing the decisions of civil servants to engage in fraudulent behaviour. Furthermore, the paper discusses the impacts of corruption on economic performance. In recent years, arguments have been advanced by some economists that corruption has negative impacts on economic growth. However, by comparing corruption and economic performance in African and Asian countries, the paper argues that the corruption-growth relationship depends not only on the scope but also on the form of corruption. This, in turn, suggests that it is possible for high levels of corruption to coexist with either high or low rates of growth, depending on how corruption is organised and how the money is used. Finally, actions to reduce corruption are discussed. The paper argues that the fight against corruption should not be seen independently from the need to reform the role of the state. Although downsizing the state and political liberalisation are desirable goals in many developing countries, they are not sufficient conditions for reducing corruption in the public sector. In some cases such moves may, in fact, lead to increased corruption (at least in the short run) by creating new opportunities.

Odd-Helge Fjeldstad

Research Professor, Coordinator: Tax and Public Finance