On the Optimal Design of Leniency Programmes

Abstract This thesis comprises of a collection of essays that aim at enhancing our under- standing of the underlying mechanics of leniency policies in antitrust. In Chapter 1, we provide a systematic overview of the most in�uential contribu- tions to the literature on collusion and leniency policies, with a focus on antitrust law. The survey elucidates the e¤ects of leniency programmes on cartel formation and cartel implementation. In Chapter 2, we provide a model to investigate the impact of a leniency pro- gramme on collusive �rms�incentives to keep or destroy hard incriminating evi- dence. We show that �rms may willfully keep the hard evidence to facilitate the implementation of the cartel. Firms are more inclined to keep the hard evidence when a leniency programme is available. Finally, �rms are more likely to destroy the hard evidence when the collusive pro�ts-�ne ratio increases. In Chapter 3, we study the strategic interaction between a cartel and an an- titrust authority whose evidence against the cartel is private information. Within the framework of a signalling game, we explore the antitrust authority�s incentives to reveal the strength of its evidence, before committing to its prosecutorial e¤ort. We show that, despite its potentially feeble evidence, the antitrust authority can exploit its informational lead and induce the cartel to self-report at an earlier stage of the prosecutorial process. The more generous the leniency programme, the easier it is to induce self-reporting by the cartel. In Chapter 4, we provide a model to characterize the optimal leniency pro- gramme when colluding �rms can invest resources to avoid detection. We show that the optimal �ne discount rate depends positively on the severity of the �ne and negatively on the probability of investigation and the cost of avoidance ac- tivities. A leniency programme that ignores �rms�e¤orts to avoid detection may result in under-deterrence