Leniency programs, antitrust enforcement and multimarket contact : three essays in industrial organization

The Layout of My Thesis This thesis contains three chapters in Industrial Organization that build on the work outlined above. The first two chapters combine leniency programs with multimarket contact and provide a thorough analysis of the potential effects of Amnesty Plus and Penalty Plus. The third chapter puts the whole discussion on leniency programs into perspective by examining other enforcement tools available to an antitrust authority. The main argument in that last chapter is that a specific instrument can only be as effective as the policy in which it is embedded. It is therefore important for an antitrust authority to know how it best accompanies the introduction or modification of a policy instrument that helps deterrence. INTRODUCTION Chapter 1 examines the efféct of Amnesty Plus and Penalty Plus on the incentives of firms to report cartel activities. The main question is whether the inclusion of these policies in a leniency program undermine the effectiveness of the latter by discouraging the firms to apply for amnesty. The model is static and focus on the ex post incentives of firms to desist from collusion. The results suggest that, because Amnesty Plus and Penalty Plus encourage the reporting of a second cartel after a first detection, a firm, anticipating this, may be reluctant to seek leniency and to report in the first place. However, the effect may also go in the opposite direction, and Amnesty Plus and Penalty Plus may encourage the simultaneous reporting of two cartels. Chapter 2 takes this idea further to the stage of cartel formation. This chapter provides a complete characterization of the potential anticompetitive and procompetitive effects of Amnesty Plus in a infinitely repeated game framework when the firms use their multimarket contact to harshen punishment. I suggest a clear-cut policy rule that prevents potential adverse effects and thereby show that, if policy makers follow this rule, a leniency program with Amnesty Plus performs better than one without. Chapter 3 characterizes the socially optimal enforcement effort of an antitrust authority and shows how this effort changes with the introduction or modification of specific policy instruments. The intuition is that the policy instrument may increase the marginal benefit of conducting investigations. If this effect is strong enough, a more rigorous detection policy becomes socially desirable.

Priming psychic and conjuring abilities of a magic demonstration influences event interpretation and random number generation biases

Magical ideation and belief in the paranormal is considered to represent a trait-like character; people either believe in it or not. Yet, anecdotes indicate that exposure to an anomalous event can turn skeptics into believers. This transformation is likely to be accompanied by altered cognitive functioning such as impaired judgments of event likelihood. Here, we investigated whether the exposure to an anomalous event changes individuals' explicit traditional (religious) and non-traditional (e.g., paranormal) beliefs as well as cognitive biases that have previously been associated with non-traditional beliefs, e.g., repetition avoidance when producing random numbers in a mental dice task. In a classroom, 91 students saw a magic demonstration after their psychology lecture. Before the demonstration, half of the students were told that the performance was done respectively by a conjuror (magician group) or a psychic (psychic group). The instruction influenced participants' explanations of the anomalous event. Participants in the magician, as compared to the psychic group, were more likely to explain the event through conjuring abilities while the reverse was true for psychic abilities. Moreover, these explanations correlated positively with their prior traditional and non-traditional beliefs. Finally, we observed that the psychic group showed more repetition avoidance than the magician group, and this effect remained the same regardless of whether assessed before or after the magic demonstration. We conclude that pre-existing beliefs and contextual suggestions both influence people's interpretations of anomalous events and associated cognitive biases. Beliefs and associated cognitive biases are likely flexible well into adulthood and change with actual life events.

Demonstration of the effectiveness of zinc in diarrhoea of children living in Switzerland.

We designed a double-blinded randomized clinical trial of zinc (10 or 20 mg of zinc sulphate for 2-5 month-old or 6-59 month-old children, respectively, during 10 days) vs. placebo in otherwise healthy children aged 2 months to 5 years who presented with acute diarrhoea (i.e. ≥3 stools/day for less than 72 h). Eighty-seven patients (median age 14 months; range 3.1-58.3) were analysed in an intention-to-treat approach. Forty-two patients took zinc and 45 placebo. There was no difference in the duration nor in the frequency of diarrhoea, but only 5% of the zinc group still had diarrhoea at 120 h of treatment compared to 20% in the placebo group (P = 0.05). Thirty-one patients (13 zinc and 18 placebo) were available for per-protocol analyses. The median (IQR) duration of diarrhoea in zinc-treated patients was 47.5 h (18.3-72) and differed significantly from the placebo group (median 76.3; IQR 52.8-137) (P = 0.03). The frequency of diarrhoea was also lower in the zinc group (P = 0.02). CONCLUSION: zinc treatment decreases the frequency and severity of diarrhoea in children aged 2 months to 5 years living in Switzerland. However, the intention-to-treat analysis reveals compliance issues that question the proper duration of treatment and the choice of optimal pharmaceutical formulation.