Do Indirect Questions Reduce Lying about Corruption? Evidence from a Quasi-Field Experiment

Recent studies have found reticent managers are less likely to report corruption than are non-reticent managers. We confirm this using new data from Bangladesh and Sri Lanka. We find reticence greatly affects estimates of corruption for measures based on both direct and indirect questions. We also find reticence affects response rates. Surprisingly, reticent managers were less likely to refuse to answer questions on corruption than non-reticent managers, possibly because reticent managers believe that refusing to answer seems like a tacit admission of guilt. Throughout the analysis, we control for the potential endogeneity of the reticence measure.