On the effects of repeated tax amnesties

Abstract

We examine empirically the effect of tax amnesties on long term tax collection when such amnesties are used by a government as a regular source of revenue. We use data from the Tucuman province (Argentina) to test the main hypothesis of the model, namely, that amnesties lower the government’s revenue, as they reduce the penalties and make evasion more profitable. We find, however, that amnesties do not affect the long-term revenue. The other main result is in line with the theoretical predictions: the increase in short-run revenue is temporary and only accelerates the collection of the taxes but does not increase the amount collected. Thus, we conclude that amnesties were used only to obtain a short-run surge in revenue and to avoid more fundamental tax reforms.