Unbundling Private and Public Sector Corruption: Insights Based on Two New Indicators

Abstract

The literature on the economics of corruption has made substantial progress in recent years. Yet most research still treats corruption as a homogenous phenomenon. This article aims to advance the study of different types of corruption by introducing two indicators that measure corruption separately for the public sector and for the private sector. One of these indicators is the first to distinctly capture private-sector corruption in a large cross-section of countries. We demonstrate the usefulness of these new indicators in two empirical applications.Firstly, we study the determinants of sector-specific corruption levels. Our results show how the causes of corruption differ between the public and private spheres. When we reduce a general congruent model including a wide range of variables to a more specific model, we identify the quality of the bureaucracy and the share of Protestants in the population as robust determinants of public-sector corruption. In contrast, none of the typical explanations of public corruption are robustly related to private-sector corruption levels.Secondly, we replicate an influential study by Fisman and Miguel (2007) and show that only public-sector corruption, and not private-sector corruption, is linked to the unlawful behavior of a country’s diplomats abroad. This suggests the necessity of a more moderate interpretation of the importance of corruption cultures than that proposed by Fisman and Miguel.