Prof. Dr. Catherine Roux, Martina Bossard

Industrieökonomie

Research Topic
Merger control aims to prevent mergers that harm competition. When markets have many competitors, mergers are usually not challenged because enough firms remain in the market after the merger to keep competition strong. However, even
in markets with many firms, problems can arise if a merger eliminates a maverick. A maverick is a firm that disrupts coordination among competitors. Losing such a firm in the post-merger market can make it easier for the remaining firms to coordinate on prices (tacitly collude), leading to higher prices and lower consumer welfare. Both the EU and US merger guidelines recognize this issue, noting that eliminating a maverick is a special circumstance where a merger can increase the risk of collusion, even in markets with many competitors. The maverick concept has appeared in several US cases, such as mergers involving Southwest Airlines, TaxACT, and T-
Mobile but it is rarely applied in EU merger decisions (Bromfield and Olczak, 2018). Critics question whether competition authorities should devote resources to identify when a merger eliminates a maverick. This project studies the strategic role of mergers eliminating mavericks and examines whether there is evidence that such mergers lead to more collusion.

Description of the Problem
The central problem is that it remains unclear whether and why mergers that eliminate mavericks actually increase coordination and should thus be treated as a distinct concern in merger control. In theory, collusion can only occur when firms interact repeatedly over time. And even then, collusion may only arise when firms are sufficiently patient to value future profits more than immediate gains. Because multiple equilibria can exist (see Folk Theorem of Fudenberg and Maskin, 1986), equilibrium selection becomes crucial. When firms are uncertain whether their competitors will collude, this strategic uncertainty makes collusion even harder to sustain (Harsanyi and Selten, 1988; Dal Bó and Fréchette, 2011). Building on this literature, Andres (2024) shows that communication between firms can reduce uncertainty and make collusion more stable. I apply Andres’ (2024) idea to the merger context. I argue that mergers influence firms’ beliefs about their competitors’ willingness to collude. A merger can thus act as a signal of collusive intent by reducing uncertainty, which makes collusion easier to sustain. Mergers that eliminate mavericks may send an even stronger signal than mergers alone, as the fi rm that previously hindered coordination disappears. My laboratory experiment tests this idea, providing evidence on whether, in practice, the maverick concept should receive greater emphasis (as in the US) or less (as in the EU).

Objectives
The goal of this project is to examine how mergers that eliminate a maverick firm affect beliefs and prices. I explore whether such mergers act as a collusive signal. My research question is: Can mergers serve as an equilibrium selection device by reducing strategic uncertainty, especially when a maverick is eliminated? I study (i) merger choice: What is the effect of a merger on beliefs? (ii) market history (presence of a maverick): What is the effect of a maverick elimination on beliefs, and the interaction of merger choice and market history on (iii) beliefs and (iv) prices.

Importance, Usefulness and Novelty of the Project
This paper has direct policy relevance and practical usefulness in the competition policy debate about the importance of the maverick concept. Theoretically, I build on established results for infinitely repeated homogeneous-goods competition à la Bertrand (1883) and model the maverick as a low-cost firm following Ivaldi et al. (2007). I then contribute to the literature through a novel application of Andres’ (2024) recent equilibrium selection approach to a merger context.
I also provide a laboratory test of the theory. Unlike earlier laboratory experiments that study either exogenously imposed (e.g., Fonseca and Normann, 2008) or endogenously chosen mergers (e.g., Lindqvist and Stennek, 2005), my design uses both to show how merger choice shapes beliefs and collusion. Earlier maverick experiments exist (Engel and Ockenfels, 2013; Darai et al., 2019), but they rely on endogenous treatment selection. The novelty of my design is to exogenously assign the maverick role, to incorporate merger choice and its interaction with the elimination of a maverick, and to study beliefs as a coordination mechanism. My project offers novel insights into the anti-competitive effects of mergers eliminating mavericks. Beyond its policy relevance, the project aligns with the topics discussed and taught in the Summer School courses at the WWZ.