Decentralisation, Corruption and Economic Growth: A Macroeconomic Perspective

This thesis represents a contribution to the literature on the relationship betweendecentralisation, corruption, and economic growth. This relationship is analysed both theoretically and empirically.The first chapter employs a dynamic general equilibrium model to gain further insights into the effects of decentralisation on the structure of corruption. The results suggest that decentralisation, by bringing the people closer to government, can enable corrupt local government officials to internalise the effects of their behaviour. It thereby generates an incentive for officials to moderate their bribe demands. This has positive effects for investment and economic growth.The second chapter examines a potential trade-off that may occur when countriesembark on a program of decentralisation. On the one hand decentralisation mayimprove the information problems that plague overly centralised governments, but at the same time it can potentially lead to a loss of control as discretionary power isgranted to local officials without implementing the required accountability mechanisms.The results of the analysis suggest that while decentralisation can potentially reducecorruption an aid economic performance in the long run, it may inevitable lead toincreased corruption in the short-run. A key idea is that extra care must be taken tointroducing accountability structures at the local level, but that these will likely taketime before becoming effective, so that in the near term corruption may increase.In the third chapter the relationship between decentralisation, corruption and economic growth is analysed empirically, using panel data techniques. While previous studies have looked at the relationship between decentralisation and corruption, or between decentralisation and growth, or between corruption and growth, few have looked at the joint relationship between the three. Moreover, previous studies often suffer from endogeneity problems. To overcome this, the Generalised Method of Moments technique is employed. It is shown that, while there is evidence that corruption hampers economic growth, the effects of decentralisation are ambiguous. The chapter highlights the inherent difficulties in analysing the effects of decentralisation, which is a complex and multifaceted concept that is impossible to fully capture in the data.Overall, the thesis provides support for the idea that decentralisation can potentiallylead to beneficial outcomes, both in terms so of combating corruption and in widereconomic terms. However, it also suggest that care must be taken when mplementing reforms as these beneficial outcomes a far from certain.