Nudging taxpayer registration? Field experimental evidence on backfiring incentives

Governments in Latin America raise little revenue from property taxation, despite arguments for its efficiency and equity. Adequate registry information would support consistent collection, but registries are costly to establish and maintain. Compared to tax collection, field experimental evidence on low-cost interventions in this area is scarce. This paper provides the first such evidence for online tax registration. The municipality of Fortaleza, Brazil, randomized 163K property taxpayers into three groups. The first group represents the status quo that did not receive a particular treatment. To the second group, the tax administration sent an e-mail asking for registration; to the third group, an e-mail that additionally offered a lottery reward for successful registration. Relative to the first group, both e-mails increased registration, especially among compliant taxpayers, men, intermediate age groups and intermediate property values. But adding the lottery incentive had a negative effect on registration. We hypothesize that this backfiring effect relates to an inadvertent signal about non-enforcement. Additional evidence from a post-experimental survey suggests that for taxpayers in the lottery group, norm compliance and the usefulness of the online registry were less important reasons to register. In sum, the results suggest that the intervention prompted parts of the population to register and that monetary incentives may be counterproductive.