A General Test of Gaming

An important lesson from the incentive literature is that explicit incentives may elicit dysfunctional and unintended responses, also known as gaming responses. The existence of these responses, however, is difficult to demonstrate in practice because this behavior is typically hidden from the researcher. We present a simple model showing that one can identify gaming by estimating the correlation between a performance measure and the true goal of the organization before and after the measure has been activated. Our hypothesis is that gaming takes place if this correlation decreases with activation. Using data from a public sector organization, we find evidence consistent with our hypothesis. We draw implications for the selection of performance measures.