Diffuse Bunching with Frictions: Theory and Estimation

We wish to acknowledge the National Treasury of South Africa for providing us with access to anonymized tax administrative data. We thank Analytics at Wharton and the Penn Wharton Budget Model for funding support. The views expressed in this paper are our own and do not necessarily reflect the views of the National Treasury of South Africa. We are grateful to Wian Boonzaaier, Ana Gamarra Rondinel, Henrik Kleven, Dylan Moore, Jacob Mortenson, Alex Rees-Jones, Juhana Siljander, Joel Slemrod, Jakob Søgaard, David Thesmar, Andrew Whitten, Eric Zwick, and seminar participants at the University of Pretoria, Economic Research South Africa (ERSA), the European Bank for Reconstruction and Development, Imperial College, IIPF 2023, LAGV 2021, LMU Munich, CREST, the NBER Public Economics Meetings, the Toulouse School of Economics, University of Cape Town, University of Michigan, and the South African Revenue Services for helpful comments and to Michael Partridge, Afras Sial, Zamir Ticknor, and Laila Voss for excellent research assistance. All errors are our own. This paper subsumes and replaces the working paper titled “Do Firms Have a Preference for Paying Exactly Zero Tax?” Computational code for the uniform sparsity model is available at https://github.com/bblockwood/DiffuseBunching. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research.

• June 20, 2024