Abstract

We explore the relationship between experiences of corruption and support for economic reform in sub-Saharan Africa. We find that the relationship varies across three rounds of the Afrobarometer survey. Examining each round separately, we find that in the first round the local intensity of bribery is correlated with support for reform. In the second round an individual’s own experience of bribery matters, while in the third round neither variable is important. Estimating our model on pooled data suggests that an individual’s own experience of corruption is associated with less support for reform on average. However, we present evidence that this association is only present in the second round. These findings point to a changing relationship which may reflect rapid development in the region, including economic growth, inward investment, and the diffusion of technology. Our key contribution is to demonstrate that policy recommendations based on an analysis of one round of data or pooled data may be misleading.