2-Stage Enforcement and Regulatory Polarisation: a Simple Model with Application to the USEPA

We present a simple, positive model in which regulatory enforcement is a 2-stage process. In the first stage firms are subject to random inspection. The inspection yields a noisy signal of the firm's true performance. Only if the signal exceeds some critical or "trigger" value is the firm subject to the second stage of enforcement, the audit. Analysis of the 2-stage process reveals some surprising comparative static results. The key result, referred to as regulatory polarisation, is the divergence of response to a tightening of the trigger. Specifically, while non-serious violators improve their performance as a result of a tighter trigger, the performance of serious violators degrades: The good get better whilst the bad get worse. The impact on total system performance is ambiguous and depends upon the initial distribution of firm types. We also analyse the implications of improvements in the accuracy of the inspection technology in such a model, arriving at some unconventional results. Whilst the model is motivated by institutional analysis of the USEPA it is argued to be applicable in a variety of other enforcement contexts.