The project aims at studying how the behaviour of firms, in particular their willingness to enter collusive arrangements, is affected by antitrust legislation and government antitrust action. The starting point is the hypothesis that antitrust policies hamper communication and other information-processing activities in which the colluding firms need to engage. The firms' choice of strategy will be determined by a trade-off between the potential benefits from participating in a cartel and the costs linked to the risk of detection. The project includes both theoretical and empirical, mainly experimental investigations.

The theoretical part builds on a new approach to the analysis of cartels that aims at handling certain shortcomings of the predominant repeated-game models. The benefits of the new approach consist in a more realistic view of the agents' cognitive limitations and a greater chance of arriving at determinate predictions.

The empirical part of the project aims at testing the conclusions of the theoretical analysis and to study collusive behaviour experimentally. In particular, the experiments will investigate how willingness to communicate is affected by the presence of a certain risk of detection and punishment.