Don’t blame the messenger. The Delivery method of a message matters

Abstract

Sending messages and providing information to individuals tends to affect their behavior (e.g., reduce energy consumption, increase donations, reduce tax evasion). Most studies to date have concentrated on evaluating the effect of different types of messages but they have avoided discussing if the method to communicate those messages could affect the effectiveness of the intervention. This study shows that the effectiveness of a message is not independent of the communication method used. We conducted a field experiment in Colombia that varies the way the National Tax Agency contacts delinquent taxpayers. More than 20,000 taxpayers were randomly assigned to a control or one of three delivery mechanisms (letter, email, and personal visit). Conditional on delivery of the treatment, a personal visit is more effective than an email, and both are more effective than a letter (the more traditional method used by tax administrations and researchers worldwide). The findings show that identifying the mechanisms through which policies are informed and publicized should be fully incorporated in the literature to assess the effectiveness of an intervention. Tax administrations should explicitly evaluate what is the right vector of communication methods that maximizes their policy objectives.