The effect of communication channels on dishonest behavior

• The effect of communication channels on dishonest behavior is experimentally tested.
• Dishonest behavior is prevalent in all treatment variations.
• Extreme dishonesty is more pronounced in increasingly distant communication.
• Partial dishonesty is more pronounced in decreasingly distant communication.

The present research investigates the effects of various communication channels on dishonest behavior. We rely on a simple truth-telling experiment (i.e., a repeated coin-flip) and let subjects report their outcome through communication channels that differ in distance and anonymity (face-to-face, in-lab telephone, in-lab web-form, and home web-form). We find dishonest behavior across all communication channels, with important treatment differences. Reporting of extreme outcomes that maximize payoff increases in distance and anonymity. To the contrary, partial lying decreases in distance and anonymity. Furthermore, we find gender to moderate the effects and women tend to drive these results. The findings have important implications for the design of real-world communication structures that are relevant when honest reporting is particularly relevant, for example in insurance claims, income reports for tax purposes, or applicant screenings in labor markets.