Revisiting the impact of corruption on income inequality worldwide

Abstract The relationship between corruption and income inequality has been widely studied, but there is no consensus on whether corruption increases or reduces inequality. We conduct an extreme bounds analysis (EBA) to test the robustness of the explanatory variables proposed in the literature. Using a sample of up to 150 countries, with data mostly going back to 1980, we find that corruption does not appear to have a clear positive effect on inequality or may even reduce it. Also, contrary to what is sometimes suggested in the literature, the results do not support an inverted U‐shaped effect of corruption on income distribution. A more important role in explaining income distribution seems to be played by the level of financial development, the old‐age dependency ratio, the unemployment rate, the capital stock to GDP ratio and the population growth rate. These are often found to be significant drivers of inequality, regardless of the set of control variables and the definition of corruption used.