Browse by title
Edited by Maureen McKelvey and Jun Jin
Elizabeth Schroeder, Carol Horton Tremblay and Victor J. Tremblay
Martin Kaae Jensen
Natalia Fabra and Massimo Motta
Mergers can give rise to two types of anticompetitive effects: unilateral effects and coordinated effects. The latter arise if after merger firms can increase their market power by coordinating their actions. In this chapter we explain what coordinated effects are and how they can be identified. As building blocks for the analysis of coordinated effects in mergers, we review the economic meaning of collusion, and assess the factors that allow firms to reach and enforce collusive outcomes. We also review some approaches for quantifying coordinated effects, and provide an overview of the use of coordinated effects in European merger control.
Catarina Marvão and Giancarlo Spagnolo
Cartels remain widespread and constitute a major problem for society. Leniency policies reduce or cancel the sanctions for the first firm(s) that self-report being part of a cartel and have become the main enforcement instrument used by competition authorities around the world in their fight against cartels. Such policies have been shown to be a powerful tool in inducing firms to self-report or cooperate with a cartel investigation in exchange for a reduction in sanctions. Since they reduce sanctions for successful leniency applicants, these programs may also be abused to generate many successful convictions for the competition authority at the expense of reduced cartel deterrence and social welfare. Hence, it is vital for competition authorities and society to understand how leniency programs affect firms’ incentives, in order to optimize their design and administration. A rich theoretical, empirical and experimental economic literature has developed in the last two decades to meet the challenge. In this chapter, we review some of the key studies that have been undertaken to date, with emphasis on more recent contributions and without claiming to be exhaustive (we apologize in advance to the authors of papers we could not discuss), highlighting and comparing the main results, and setting out their limitations. We conclude with a general assessment and an agenda for future research on this topic at the core of competition policy.
Marco A. Marini
This survey introduces a number of game-theoretic tools to model collusive agreements among firms in vertically differentiated markets. I first review some classical literature on collusion between two firms producing goods of exogenous different qualities. I then extend the analysis to an n-firm vertically differentiated market to study the incentive to form either a whole market alliance or partial alliances made of subsets of consecutive firms in order to collude in prices. Within this framework I explore the price behaviour of groups of colluding firms and their incentive to either prune or proliferate their products. It is shown that a selective pruning within the cartel always occurs. Moreover, by associating a partition function game to the n-firm vertically differentiated market, it can be shown that a sufficient condition for the cooperative (or coalitional) stability of the whole industry cartel is the equidistance of firms’ products along the quality spectrum. Without this property, and in presence of large quality differences, collusive agreements easily lose their stability. In addition, introducing a standard infinitely repeated game approach, I show that an increase in the number of firms in the market may have contradictory effects on the incentive of firms to collude: it can make collusion easier for bottom and intermediate firms and harder for the top-quality firm. Finally, by means of a three-firm example, I consider the case in which alliances can set endogenously qualities, prices and number of variants on sale. I show that, in every formed coalition, (i) market pruning dominates product proliferation and (ii) partial cartelization always arises in equilibrium, with the bottom-quality firm always belonging to the alliance.