Endogenous Crackdowns: Theory and Experimental Evidence

We present a simple model to analyze the efficacy of endogenous crackdowns, i.e., sudden and dramatic increases of audit probability triggered by a low level of compliance. We test this model experimentally. Our results show that: (a) compliance reacts quickly to the occurrence of crackdowns; (b) subjects report more than half of their income even during non-crackdown periods; (c) announcements of crackdown increase significantly tax compliance both when crackdowns are pre-announced and when they are announced ex post; (d) subjects are able to coordinate quickly to end crackdowns. Our results have important implications for policy-makers in terms of designing more effective crackdown policies.