Cartels can severely harm social welfare. Competition authorities introduced leniency rules to destabilize existing cartels and hinder the formation of new ones. Empirically, it is difficult to judge the success of these measures because functioning cartels are unobservable. Existing experimental studies confirm that a leniency rule indeed reduces cartelization. We extend these studies by having a participant in the role of the competition authority actively participating in the experiment. Based on chat communication content and price setting behavior, this authority judges whether firms formed a cartel and decides on fines in real time. We find that a leniency rule does not affect cartelization in this setup.