Modelling Corruption Perceptions: Evidence from Eastern Europe and Central Asian Countries

This work proposes a multidimensional framework that is based on a latent class model to identify various types of corruption and to outline their importance. A dataset of Eastern European and Central Asian countries is used to identify four groups of corrupt activities, which go beyond the usual classification of corruption into administrative and political corruption. Our estimates are validated by means of a direct administrative corruption index that is derived from the same dataset and also by a comparison with the corruption perception rankings that are published by Transparency International. The potential of the proposed approach is illustrated with an application to the relationship between firms’ competitiveness and the latent classes of corruption that we have identified.