Tax Compliance in Rwanda: Evidence from a Message Field Experiment

Abstract

In early 2016, the Rwanda Revenue Authority sent 9,000 taxpayers messages aimed at encouraging compliance. Each taxpayer was randomly allocated to a treatment group or a control group that received no message. Treatment messages varied in terms of content (deterrence, fiscal exchange, or reminder) and delivery method (letter, email, or SMS). Our randomized controlled trial evaluates the impact of this intervention on compliance behavior. We find that friendly approaches—especially a simple reminder of deadlines—are more effective than deterrence. We also provide suggestive evidence that deterrence messages produce a backfiring effect among higher-income taxpayers, while they have the expected effect for the rest.