Abstract
This paper tests nudges delivered via WhatsApp to encourage timely annual tax returns among 24,887 individual non-employee taxpayers in six tax offices. We compared a standard email to three treatments delivered via a verified WhatsApp business account: (i) a standard message, (ii) a deterrence message, and (iii) a reciprocity message. The pre-registered design permits separating the effects of the mode of delivery (WhatsApp vs. email) and behavioral messaging. 37.7% of the control group submit their annual tax return on time. Based on WhatsApp status reports, we find that 44.2% of treated taxpayers received and read the messages. The standard reminder via WhatsApp increased tax return rates by 3.3 percentage points. The addition of the reciprocity message did not yield further improvement. Instead, we find that the deterrence message increased compliance by an additional 1.9 percentage points. We investigate heterogeneity and sensitivity to deviations from the pre-registered design including implementation.
Original language | English |
---|---|
Pages (from-to) | 1-28 |
Number of pages | 28 |
Journal | Bulletin of Indonesian Economic Studies |
DOIs | |
Publication status | Published - 18 Dec 2024 |