Spatial Tax Enforcement Spillovers: Evidence from South Africa

The purpose of this paper is to test for spatial enforcement spillovers. Empirical testing ground is the enforcement of business taxes in South Africa. The analysis relies on the population of business tax returns for the years 2009 to 2014 and data on all business taxpayer audits by the South African Revenue Services during that time period. The results suggest that audits significantly raise the tax reporting of non-audited neighboring firms. While the observed spillovers decline in geographic distance to the audited entity and are short-run in nature, the implied aggregate revenue gains are nonnegligible. Additional analysis shows that the effect is driven by audit cases, where audited firms do not experience an upward revision in their tax owed in the course of the audit. This suggests that the observed effect roots in communication among taxpayers and is not driven by audit-related cost shocks to business partners.