Flexibility and Collusion with Imperfect Monitoring (original) (raw)

Frequency of interaction, communication and collusion: an experiment

Economic Theory, 2018

The frequency of interaction facilitates collusion by reducing gains from defection. Theory has shown that under imperfect monitoring flexibility may hinder cooperation by inducing punishment after too few noisy signals, making collusion impossible in many environments (Sannikov and Skrzypacsz, AER 2007). The interplay of these forces should generate an inverse U-shaped effect of flexibility on collusion. We test for the first time these theoretical predictionscentral to antitrust policy-in a laboratory experiment featuring an indefinitely repeated Cournot duopoly, with different degrees of flexibility. Results turn out to depend crucially on whether subjects can communicate with each other at the beginning of a supergame (explicit collusion) or not (tacit collusion). Without communication, the incidence of collusion is low throughout and not significantly related to flexibility; when subjects are allowed to communicate, collusion is significantly more frequent in the treatment with intermediate flexibility than in the treatments with low or high flexibility.

Flexibility, Communication and Cooperation with Imperfect Monitoring

Flexibility-the possibility to react swiftly to others' choicesfacilitates cooperation by reducing the gains from defection. With imperfect monitoring, however, flexibility may also hinder cooperation by inducing punishment after too few noisy signals. In theory, the interplay of these forces should generate an inverse U-shaped effect of flexibility on cooperation. To test this subtle prediction, we implement a repeated Cournot game under imperfect monitoring in the lab and change the delay with which subjects can react to noisy signals. Results are as subtle as the theoretical prediction, and depend crucially on subjects' ability to communicate. A content analysis of the communication confirms the behavioral relevance of the main strategic forces highlighted by the theory.

Communication, Renegotiation, and the Scope for Collusion

SSRN Electronic Journal, 2000

We study the effect of communication in an experimental game where cooperation is consistent with equilibrium play if players share an understanding that cheating will be punished. Consistent with communication acting as a coordinating device, credible preplay threats to punish cheating are the most effective message to facilitate collusion. Promises to collude also improve cooperation. Credible threats do not occur in a treatment with a limited message space that permits threats of punishment. Contrary to some theoretical predictions, renegotiation possibilities facilitate collusion. (JEL C71, C73, D83, L12)

Dynamic Competition with Irreversible Moves: Tacit Collusion (Almost) Guaranteed

We consider a new model to analyze the dynamics of competitive interactions. The players interact repeatedly, but are restricted to employ (weakly) increasingly competitive strategies (e.g. increasing quantities, decreasing prices). In a wide range of circumstances, unique equilibrium paths result, and often those imply tacit collusion (e.g. cartel capacities, monopoly prices). There are several possible motivations of irreversibilities. Physically (besides the literal motivation), the players might constrain their strategy sets themselves (which, paradoxically, can simplify tacit collusion). Technically, the assumption of irreversible moves serves similar ends as that of Markov strategies, and the induced dynamics resemble (re)negotiations.

On the Difficulty of Collusion in the Presence of a More Efficient Outsider

Journal of Institutional and Theoretical Economics, 2018

On the Difficulty of Collusion bitte auf den kommenden ungeraden Seiten und Guillaume Cheikbossian and Philippe Mahenc auf den geraden Seiten als Seitentitel setzen. We study the ability of several identical firms to collude in the presence of a more efficient firm, which does not take part in their collusive agreement. The cartel firms adopt stick-and-carrot strategies, while the efficient firm plays its one-period best-response function, regardless of the history of play. We characterize the most collusive symmetric punishment, which maximizes the scope for collusion. We then find that either a lower cost disadvantage or a smaller cartel size facilitates collusion. Finally, we compare our results with those obtained in the standard setup where all firms participate in the collusive agreement.

Sustaining collusion in markets with a general evolution of demand

2014

This paper proposes a general framework to study the sustainability of collusion in markets where demand growth (although deterministic) is not restricted to occur at a constant rate and may trigger future entry. It is shown that, typically, entry occurs later along the collusive path than along the punishment path (since profits are lower in the latter case). The possibility of delaying entry, therefore, constitutes an additional incentive for deviating just before entry is supposed to occur along the collusive path. In case discontinuing collusion does not delay entry, collusion is shown to be typically more difficult to sustain after than before entry. The proposed model encompasses and explains conflicting results derived in the extant literature under more restrictive settings, and derives some additional results. In particular, it is shown that whether collusion is more difficult before or after entry crucially depends on the magnitude of the entry costs and on the speed at wh...

Collusion with secret price cuts: an experimental investigation

Theoretical work starting with Stigler (1964) suggests that collusion may be difficult to sustain in a repeated game with secret price cuts and demand uncertainty. Compared to equilibria in games of perfect information, trigger−strategy equilibria in this context result in lower payoffs because punishments occur along the equilibrium path. We tested the theory in a series of economic experiments. Consistent with the theory, treatments with imperfect information were less collusive than treatments with perfect information. However, in the imperfect−information treatments, players seemed to settle on the static Nash outcome rather than using trigger strategies. Players did resort to punishments for undercutting in perfect−information treatments, and this sometimes led to successful collusion afterward. We would like to thank Sabine Bendanoun, Yvon Pho, and Tad Reynolds for research assistance and Doug Davis, Tom Husted, and Roger Sherman for helpful comments. Financial support was provided by a grant from the College of the Arts and Sciences at American University. Views expressed are those of the authors alone.

On Communication and Collusion†

American Economic Review, 2016

We study the role of communication within a cartel. Our analysis is carried out in Stigler’s (1964) model of repeated oligopoly with secret price cuts. Firms observe neither the prices nor the sales of their rivals. For a fixed discount factor, we identify conditions under which there are equilibria with “cheap talk” that result in near-perfect collusion, whereas all equilibria without such communication are bounded away from this outcome. In our model, communication improves monitoring and leads to higher prices and profits. (JEL C73, D43, D83, L12, L13, L25)