The Dynamic Relationship Between Corruption–Inflation: Evidence From Panel Vector Autoregression

Abstract

This paper aims to explain the mixed causality nexus between corruption and inflation. For that, we apply a panel vector autoregression model on a large sample of 180 countries over the period 1996–2014. Using two corruption indexes and subsample estimations, results provide evidence that the inflation–corruption nexus is bidirectional. The causal effect is more important from corruption to inflation. Interactions remain significant but heterogeneous across subsamples with different income levels. The corruption effect is persistent only in low–middle income economies and its adverse effect on inflation is weaker in high-income economies. The two-way relationship between inflation and corruption reflects the inability to control inflation and the situation of the poverty trap in some countries.