This paper investigates the role of targeting in the context of agri-environmental schemes involving monitoring and penalties and well suited to a geographically-based distinction between participants. By separating participants into a target and a non-target group the aim of targeting is to reduce the moral hazard problem. The paper analyses three approaches to targeting and the focus is on reducing the extent of cheating by participants in the non-target group. By complementing the adoption of targeting with appropriate adjustments to the monitoring/penalty parameters it is shown how such an approach can exploit the risk aversion of participants to completely eliminate cheating by those participants in the non-target group.