Obtaining insight into the effects of policy interventions is often a difficult matter. A new method to obtain a first insight into those effects is presented in this paper. The basis of the method is a Causal Loop Diagram to which information on causal relations and variables is added. Part of the information is expressed in qualitative terms. This Method to Analyse Relations between Variables using Enriched Loops (MARVEL) takes proposed interventions as a starting point. Interventions are interpreted as imposed changes on selected model variables representing intervention points. A new feature is that causal relations are no longer passive but active model elements. They propagate the changes through the model in a time-dispersed way. MARVEL determines how this causes (other) variables representing the modelís performance to change in the desired direction at selected moments in time. MARVEL can be used for policy development, policy analysis and policy evaluation problems.