Association for Public Policy Analysis & Management

There is a growing literature on mobility of high-income people in response to tax differentials (Kleven et al. 2019), both across and within countries. While this literature has documented substantial mobility responses, an open question is the extent to which these responses reflect real mobility or merely income shifting. In this work, I exploit variation in local income tax rates across Italian municipalities and ask whether tax differentials are associated with a higher likelihood of misreporting individual fiscal residence. First, I estimate mobility responses similar in magnitude to what the literature have found for other countries. Then, to disentangle real mobility from income shifting, I construct a proxy of individual misreporting by combining two measures of residence for employees: their fiscal residence and the location of their workplace. I then regress this misreporting measure on the local tax differential as well as other individual, employer and location specific controls. Under the assumption that measurement error (e.g. due to commuting or remote working) is not correlated with local tax differentials, conditionally on controls, I will interpret any significant relationship between tax differentials and misreporting as evidence of income shifting. I will then back-of-the-envelope estimate the likelihood of a real migration response when a reported move is observed, as well as the distributional implications for local government revenues. These findings could have important implications for local public finance but also for policymakers and researchers who rely on population counts, which could be biased if systematic misreporting is at work.