Taking State-Capacity Research to the Field: Insights from Collaborations with Tax Authorities

Abstract

No modern state can exist in the long term without effective taxation. Recent research emerging from close collaboration of academics with tax authorities has shed new light on how states can build such tax capacity. Using both randomized and natural experiments, these partnerships have not only opened access to new types of data but have also stimulated new perspectives and research questions. While much of research in public finance has historically assumed that a tax in the law is a tax that is collected, exciting new research takes an empirical look inside the black box of tax administration. It addresses issues ranging from the role of information and digitalization to taxpayer behavior or to the link between taxation and citizens’ relationship to the state. This article provides a brief overview of some of this research, as well as practical advice for those interested in implementing research in partnership with tax authorities or other large public entities.

Nudging for Tax Compliance: A Meta-Analysis

“…A competing interpretation, however, is that deterrence and nondeterrence nudges are not equally effective in shifting the prior beliefs of taxpayers. According to Pomeranz and Vila-Belda (2019), it may well be that nudges implemented by tax authorities are more effective in shifting perceptions of audit probabilities than perceptions of social norms.…”

The Impact of COVID-19 on Formal Firms: Micro Tax Data Simulations across Countries

“…Balla-Elliott et al (2020) shows that firms expect demand to be a third lower than pre-COVID levels upon re-opening. 2 SeePomeranz and Vila-Belda (2019) andSlemrod (2018) for a summary of recent studies using administrative tax data.…”

Nobel Rebels in Disguise — Assessing the Rise and Rule of the Randomistas

“…Randomistas often present their quest for evidence as neutral and in everyone's best interest. For example, Naritomi et al (2020, p. 1) argue that RCTs bring academics, policymakers from governments, NGOs and international organizations together 'who share an interest in credibly estimating the causal effects of development programs' (see Duflo et al 2015b;Pomeranz and Vila-Belda 2019). Similarly, when the Innovations for Poverty Action (IPA) website states that 'When our evidence goes out into the world, it can be used by anyone,' 19 it suggests that experimental research has the status of an accessible public good that anyone can pick up and use, rather than producing extremely context-specific estimations.…”