Working Paper 215
The literature on tax administration and compliance in developing countries has been blooming recently, primarily driven by the availability of administrative data and by the increased urgency of revenue mobilisation in the context of multiple global crises. Yet, one key aspect of tax administration has received little attention in this literature so far: its interpersonal nature. In the practical reality of taxpaying processes, the experience of taxpayers is largely shaped by their interactions with the people who implement tax rules and laws: tax officials.
This paper starts filling this gap by investigating the impact of interactions between taxpayers and tax officials in Rwanda, focussing specifically on income taxpayers. We adopt a mixed methods approach to analysing taxpayer–official interactions, based on focus group discussions (FGD) with taxpayers, in-depth interviews with tax officials, and a taxpayer survey that includes a vignette experiment. The key theme emerging from FGDs and interviews is taxpayer knowledge and the related need for education: weak tax knowledge results in confusion and anxiety on the taxpayers’ side and more challenging interactions with tax officials. Tax knowledge (related to firm size) is a key source of heterogeneity in taxpayer experiences of interaction with the tax administration, alongside differences in tax officials’ characteristics, such as seniority or location (e.g. headquarters vs local branches). Our vignette-based experiment confirms that taxpayer–official interactions play a significant role in shaping compliance perceptions. In line with the qualitative results, recalling interactions based on facilitation improves trust in the tax administration and perceptions of professionalism amongst officials. On the contrary, recalling interactions focussed on enforcement deteriorates perceptions about tax morale, respect towards the tax administration, and tax officials’ aggressiveness – with backfiring effects concentrated especially amongst larger taxpayers. Finally, both treatments result in an increase in anxiety and a decrease in perceptions of client orientation. We also document heterogeneous treatment effects across relevant sub-groups of taxpayers and tax officials.
