Competing explanations for parallel conduct Lessons from the Australian detergent case (ACCC v Colgate-Palmolive)

Click here to read the full article online

Parallel conduct by competing firms can be the result of completely independent and uncontroversial behaviour, such as when all suppliers are affected by and respond unilaterally to the same increase in costs. At the other extreme, parallel conduct can be the result of interdependent and deliberately coordinated behaviour, such as when all suppliers meet in the proverbial smoke-filled room and agree to fix prices. But as economists have been telling us for decades, there is a vast middle ground, where the parallel conduct stems from some degree of interdependence and some behaviour short of the hard-core cartel described above. Understanding the reasons for the parallel behaviour and deciding what to do about it is a central element of modern antitrust law and policy. Our goal in this paper is to make some incremental progress by exploring, in the Australian context, how a firm can escape the allegation that it has been a party to a contract, arrangement or understanding even when its conduct in the marketplace appears to be roughly parallel to that of its competitors. Our vehicle for this approach is the decision of the trial judge in the laundry detergent case (ACCC v Colgate-Palmolive), upheld by the Full Court of the Federal Court of Australia, to dismiss the case brought by the Australian Competition and Consumer Commission against Cussons, the one producer of detergent that actually went to trial, determining that the ACCC had not succeeded in proving that Cussons was a party to any collusive arrangement or understanding with its competitors (Colgate and Unilever) or with one of its major customers (Woolworths). It is our hope that an analysis of the evidence in the case and the claims that were made (and either accepted or rejected) based on that evidence will assist in framing the ongoing debate about where the boundary between lawful and unlawful conduct lies (or should lie) and will highlight the kinds of economic evidence that advance the debate.