This symposium was motivated by a concern that vertical restraints dampen competition in a world with otherwise reasonably effective anti-cartel policies because cartels make more frequent use of vertical restraints than has generally been recognized. Our examination of explicit horizontal European Commission cartel cases from the 1990s found that a surprisingly large fraction (one quarter) of cartels were making use of vertical relationships to sustain, and sometimes to disguise, their collusive behavior. The European Commission was not specifically searching for evidence of vertical restraints, as documenting such behavior was not necessary to make the legal case of a violation of competition rules against horizontal price fixing or market allocation. Rather, in the process of describing the organization and functioning of each cartel, the European Commission documented the use of vertical restraints.
It is impossible to know how common these cases are because there has been no systematic collection of information on these kinds of activities, even where illegal horizontal collusion has been discovered. European competition authorities are increasingly reluctant to share details of investigations, in part because of the increased risk of individual, personal liability for anticompetitive practices, the increased risk of civil liability for firms, and sequential or differential implementation of leniency across jurisdictions. These risks may create incentives to reduce horizontal collusion, but an unintended consequence of the European competition authorities’ lack of transparency is to make the organization and mechanisms of anticompetitive behavior arising from other kinds of behavior, such as vertical restraints, more obscure.
Does the finding—that one-quarter of a sample of international cartels in the 1990s and early 2000s showed evidence of vertical relationships that support collusion—suggest that this occurs in different countries and across legal regimes? Unfortunately, the information required to establish this is not clear. We examined the records of horizontal collusion cases from six different (English-speaking) jurisdictions: Australia, Canada, Ireland, South Africa, United Kingdom, and United States. The descriptions of the cases were too limited to discern whether vertical restraints were used by cartel members to support horizontal collusion. There were suggestions of vertical restraints in a few cases, which we discuss below, but publicly available evidence was largely non-existent.
Observers often miss the role of vertical relationships because both the economic and legal frameworks encourage us to characterize interfirm relationships as either horizontal or vertical. In this article we note that horizontal and vertical relationships often work in tandem to enable anticompetitive behavior and reduce efficiency. Downstream firms may help to prevent cheating and entry or assist firms in coordinating their behavior on a collusive outcome. Downstream firms receive a share of collusive rents in return for these activities. Robert Bork famously and influentially argued that vertical restraints should be per se legal, because, as mutually agreed upon, voluntary contracts, they could not harm consumers or competition. A number of writers, including some of the contributors to this volume, have demonstrated that this is not the case theoretically or empirically. Vertical restraints are not always innocuous; sometimes they are being used to facilitate collusion.
This article focuses, through discussion of a series of cartel case studies, on the ways that distributors may facilitate collusion. We choose this focus not because we think that distributors are more likely to facilitate than undermine collusion, but rather because antitrust policy does not need to change to address cases where distributors undermine collusion. When distributors make it easier for firms to cheat on collusive arrangement or play upstream firms against one another, consumers benefit; competition policy’s laissez-faire attitude toward vertical relationships is vindicated. Policy intervention is necessary, however, when the market does not automatically move the economy toward an efficient outcome. Distributors’ potential involvement in collusion is such a case. The possibility that distributors are participating in horizontal collusion requires awareness by policymakers and a change in antitrust policy to address it.
